U.S. patent number 9,270,898 [Application Number 14/206,198] was granted by the patent office on 2016-02-23 for camera and image processing method for spectroscopic analysis of captured image.
This patent grant is currently assigned to Seiko Epson Corporation. The grantee listed for this patent is Seiko Epson Corporation. Invention is credited to Tatsuaki Funamoto.
United States Patent |
9,270,898 |
Funamoto |
February 23, 2016 |
Camera and image processing method for spectroscopic analysis of
captured image
Abstract
A spectroscopic analysis apparatus includes a light source
section that radiates light toward an object being imaged, an
imaging section that captures light reflected off the object being
imaged to acquire an image, a pixel detector that detects an
abnormal pixel in the image which is a pixel where a reflectance
ratio is greater than or equal to 1 and detects normal pixels in
the image each of which is a pixel where the reflectance ratio is
smaller than 1, and a light amount corrector that calculates a
light amount correction value based on the amounts of light at
normal pixels in a pixel area including the abnormal pixel in the
image and replaces the amount of light at the abnormal pixel with
the light amount correction value.
Inventors: |
Funamoto; Tatsuaki (Shiojiri,
JP) |
Applicant: |
Name |
City |
State |
Country |
Type |
Seiko Epson Corporation |
Tokyo |
N/A |
JP |
|
|
Assignee: |
Seiko Epson Corporation
(JP)
|
Family
ID: |
51525733 |
Appl.
No.: |
14/206,198 |
Filed: |
March 12, 2014 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20140267840 A1 |
Sep 18, 2014 |
|
Foreign Application Priority Data
|
|
|
|
|
Mar 13, 2013 [JP] |
|
|
2013-050096 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01J
3/26 (20130101); H04N 5/2176 (20130101); G01J
3/2823 (20130101); G01J 3/28 (20130101); H04N
5/243 (20130101) |
Current International
Class: |
H04N
5/243 (20060101); H04N 5/217 (20110101); G01J
3/26 (20060101); G01J 3/28 (20060101) |
Field of
Search: |
;348/241 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
2000-329617 |
|
Nov 2000 |
|
JP |
|
2006-170669 |
|
Jun 2006 |
|
JP |
|
2009-033222 |
|
Feb 2009 |
|
JP |
|
Primary Examiner: Selby; Gevell
Attorney, Agent or Firm: Harness, Dickey & Pierce,
P.L.C.
Claims
What is claimed is:
1. A camera comprising: a light source section that radiates light
toward an object being imaged; an imaging section that captures
light reflected off the object being imaged to acquire an image; a
pixel detection section that detects an abnormal pixel in the image
which is a pixel where a ratio of the amount of light at the pixel
to a reference amount of light obtained when a reference object is
irradiated with the light is greater than or equal to a
predetermined value and detects normal pixels in the image each of
which is a pixel where the ratio is smaller than the predetermined
value; a light amount correction section that calculates a light
amount correction value based on the amounts of light at normal
pixels out of the normal pixels that are separated from the
abnormal pixel in the image and located within a predetermined
distance range, wherein the light amount correction section
replaces the amount of light at the abnormal pixel with the light
amount correction value; and a spectroscopic device that selects
and separates light of a predetermined wavelength from the light
reflected off the object being imaged, wherein: the imaging section
captures the light of the wavelength selected by the spectroscopic
device to acquire the image, and the spectroscopic device is
capable of changing the wavelength to be selected.
2. The camera according to claim 1, wherein the light amount
correction section calculates the light amount correction value in
the form of the average of the amounts of light at the normal
pixels located within the predetermined distance range.
3. The camera according to claim 1, wherein the light amount
correction section calculates the light amount correction value in
the form of the median of the amounts of light at the normal pixels
located within the predetermined distance range.
4. The camera according to claim 1, wherein the light amount
correction section calculates the light amount correction value in
the form of the average of the amounts of light in the quartile
range at the normal pixels located within the predetermined
distance range.
5. The camera according to claim 1, further comprising an input
section that sets the predetermined value based on which each pixel
in the image is determined to be an abnormal pixel or a normal
pixel.
6. The camera according to claim 1, further comprising an input
section that sets the predetermined distance range that defines a
range within which the normal pixels used to correct the amount of
light at the abnormal pixel are located.
7. The camera according to claim 1, wherein the spectroscopic
device is a wavelength tunable Fabry-Perot etalon.
8. An image processing method in a camera including a light source
section that radiates light toward an object being imaged, the
method comprising: selecting and separating light of a
predetermined wavelength from the light reflected off the object
being imaged; capturing the light of the wavelength selected to
acquire the image; changing the wavelength to be selected;
detecting an abnormal pixel in the image which is a pixel where a
ratio of the amount of light at the pixel to a reference amount of
light obtained when a reference object is irradiated with the light
is greater than or equal to a predetermined value and detecting
normal pixels in the image each of which is a pixel where the ratio
is smaller than the predetermined value; and correcting a light
amount by calculating a light amount correction value for
correcting the abnormal pixel based on the amounts of light at
normal pixels out of the normal pixels that are located in a
predetermined pixel area around the abnormal pixel in the image and
replacing the amount of light at the abnormal pixel with the light
amount correction value.
9. A camera comprising: a light source section that radiates light
toward an object being imaged; an imaging section that captures
light reflected off the object being imaged to acquire an image; a
pixel detection section that detects an abnormal pixel in the image
which is a pixel where a ratio of the amount of light at the pixel
to a reference amount of light obtained when a reference object is
irradiated with the light is greater than or equal to a
predetermined value and detects normal pixels in the image each of
which is a pixel where the ratio is smaller than the predetermined
value; a light amount correction section that calculates a light
amount correction value based on the amounts of light at normal
pixels out of the normal pixels that are separated from the
abnormal pixel in the image and located within a predetermined
distance range, wherein the light amount correction section
replaces the amount of light at the abnormal pixel with the light
amount correction value; and an input section that sets the
predetermined distance range that defines a range within which the
normal pixels used to correct the amount of light at the abnormal
pixel are located.
10. An image processing method in a camera including a light source
section that radiates light toward an object being imaged and an
imaging section that captures light reflected off the object being
imaged to acquire an image, the method comprising: detecting an
abnormal pixel in the image which is a pixel where a ratio of the
amount of light at the pixel to a reference amount of light
obtained when a reference object is irradiated with the light is
greater than or equal to a predetermined value and detecting normal
pixels in the image each of which is a pixel where the ratio is
smaller than the predetermined value; setting a predetermined
distance range that defines a range within which the normal pixels
used to correct the amount of light at the abnormal pixel are
located; and correcting a light amount by calculating a light
amount correction value for correcting the abnormal pixel based on
the amounts of light at normal pixels out of the normal pixels that
are located within the predetermined distance range from the
abnormal pixel in the image and replacing the amount of light at
the abnormal pixel with the light amount correction value.
Description
BACKGROUND
1. Technical Field
The present invention relates to a camera and an image processing
method.
2. Related Art
There is a known apparatus of related art that radiates light
toward an object being imaged and captures light reflected off the
object being imaged to produce a captured image (see
JP-A-2009-33222, for example).
The imaging apparatus (spectroscopic camera) described in
JP-A-2009-33222 causes light from the object to be incident on a
Fabry-Perot interference filter and allows an image sensor to
receive light having passed through the Fabry-Perot interference
filter for acquisition of a spectroscopic image.
A spectroscopic camera using a Fabry-Perot interference filter has
an advantage of being compact and lightweight, as described in
JP-A-2009-33222. On the other hand, to acquire a spectroscopic
image based on a sufficient amount of near infrared light, a near
infrared light source needs to be provided in an imaging apparatus
body. However, providing such a light source in the spectroscopic
camera, which is compact as described above, results in a short
distance between the light source and an imaging lens. In this
case, light specularly reflected off the surface of an object being
imaged enters the imaging lens, undesirably resulting in abnormal
brightness of part of a spectroscopic image.
SUMMARY
An advantage of some aspects of the invention is to provide a
camera capable of acquiring a high-precision image even when light
from a light source is specularly reflected off the surface of an
object being imaged and also provide an image processing
method.
An aspect of the invention is directed to a camera including a
light source section that radiates light toward an object being
imaged, an imaging section that captures light reflected off the
object being imaged to acquire an image, a pixel detection section
that detects an abnormal pixel in the image which is a pixel where
the ratio of the amount of light at the pixel to a reference amount
of light obtained when a reference object is irradiated with the
light is greater than or equal to a predetermined value and detects
normal pixels in the image each of which is a pixel where the ratio
is smaller than the predetermined value, and a light amount
correction section that calculates a light amount correction value
based on the amounts of light at normal pixels out of the normal
pixels that are separated from the abnormal pixel in the image but
located within a predetermined distance range and replaces the
amount of light at the abnormal pixel with the light amount
correction value.
The reference object in the aspect of the invention is, for
example, a reference white plate or any object having a perfect
diffusing surface or a quasi-perfect diffusing surface. Consider a
case where a perfect diffusing surface is irradiated with light and
the amount of light reflected off the perfect diffusing surface is
used as a reference amount of light. The ratio of the amount of
light at each pixel in a captured image to the reference amount of
light is a reflectance ratio with respect to the perfect diffusing
surface. When specular reflection occurs at a portion of the
surface of the object being imaged, the portion has reflectance
greater than that of the perfect diffusing surface, and the
reflectance ratio is therefore greater than "1". The pixel
detection section can therefore detect an abnormal pixel
corresponding to the specular reflection portion and a normal pixel
corresponding to a diffuse reflection portion.
The reference amount of light is not limited to the amount of light
reflected off a perfect diffusing surface. For example, the surface
of the reference object may absorb or otherwise interact with part
of light incident thereon. In this case, the reflectance has a
finite value lower than 100% (99%, for example). When the amount of
light reflected off such a reference object is used as the
reference amount of light, the pixel detection section can detect a
pixel where the reflectance ratio is greater than a predetermined
value smaller than 1 (0.99, for example) as an abnormal pixel
corresponding to specular reflection.
In the aspect of the invention, the pixel detection section detects
an abnormal pixel and normal pixels in an image, and the light
amount correction section calculates a light amount correction
value based on the amounts of normal pixels around the abnormal
pixel and replaces the amount of light at the abnormal pixel with
the light amount correction value. The amount of light at the
abnormal pixel can thus be replaced with an appropriate light
amount value, whereby an image having no pixel showing an abnormal
amount of light corresponding to a specular reflection portion can
be acquired.
In the camera according to the aspect of the invention, it is
preferable that the camera further includes a spectroscopic device
that selects and separates light of a predetermined wavelength from
the light reflected off the object being imaged, and the imaging
section captures the light of the wavelength selected by the
spectroscopic device to acquire the image.
In the aspect of the invention with the configuration described
above, as the image to be acquired, a spectroscopic image produced
by capturing light of a predetermined wavelength separated by the
spectroscopic device is acquired. In this configuration, the amount
of light at an abnormal pixel corresponding to a specular
reflection portion in the spectroscopic image can be corrected to
the light amount correction value calculated based on the amount of
light at a normal pixel, whereby a high-precision spectroscopic
image can be acquired.
In the camera according to the aspect of the invention, it is
preferable that the light amount correction section calculates the
light amount correction value in the form of the average of the
amounts of light at the normal pixels located within the
predetermined distance range.
In the aspect of the invention with the configuration described
above, since the light amount correction value is the average of
the amounts of light at normal pixels, the light amount correction
value can be readily calculated.
In the camera according to the aspect of the invention, it is
preferable that the light amount correction section calculates the
light amount correction value in the form of the median of the
amounts of light at the normal pixels located within the
predetermined distance range.
In the aspect of the invention with the configuration described
above, the median of the amounts of light at normal pixels is
calculated as the light amount correction value. When the amounts
of light at the pixels in the image vary from each other, for
example, when an edge portion is present in a pixel area, using the
average as the light amount correction value possibly results in a
large error. In contrast, when the median is used as the light
amount correction value, the amount of light that belongs to a
large population can be used in the case where the amounts of light
vary from each other in the pixel area, whereby correction using a
normal amount of light is more likely to be made.
In the camera according to the aspect of the invention, it is
preferable that the light amount correction section calculates the
light amount correction value in the form of the average of the
amounts of light in the quartile range at the normal pixels located
within the predetermined distance range.
In the aspect of the invention with the configuration described
above, the average of the amounts of light at normal pixels in the
quartile range is calculated as the light amount correction value.
In this case, any pixel located within the pixel area but showing
an amount of light that greatly deviates from those at other normal
pixels can be excluded, whereby the amount of light at an abnormal
pixel can be likely to be corrected by using a more appropriate
amount of light.
In the camera according to the aspect of the invention, it is
preferable that the camera further includes an input section that
sets the predetermined value based on which each pixel in the image
is determined to be an abnormal pixel or a normal pixel.
In the aspect of the invention with the configuration described
above, the camera further includes an input section that sets the
predetermined value, and a user can, for example, operate the input
section to set the predetermined value. Further, the input section
may be configured, for example, to automatically change (reduce,
for example) the predetermined value when the number of abnormal
pixels in an acquired image is greater than or equal to a
predetermined upper limit.
In this configuration, for example, when too many abnormal pixels
in a captured image lower the precision in the light amount
correction based on normal pixels, abnormal pixel detection
sensitivity can be lowered for more appropriate light amount
correction.
In the camera according to the aspect of the invention, it is
preferable that the camera further includes an input section that
sets the predetermined distance range that defines a range within
which the normal pixels used to correct the amount of light at the
abnormal pixel are located.
In the aspect of the invention with the configuration described
above, the camera further includes an input section that set the
predetermined distance, and the user can, for example, operate the
input section to set the predetermined distance. Further, the input
section may be configured to automatically change the predetermined
distance in accordance with the number of abnormal pixels located
around the abnormal pixel in an acquired image, the position of an
edge portion where the amount of light greatly changes, or other
factors.
In this configuration, for example, when a predetermined distance
set based on user's operation is used, the user can check an image
and then set an area (predetermined distance described above) in
which a small number of abnormal pixels are present. Further, when
such an area is automatically set in accordance with the number of
normal pixels around an abnormal pixel, and the number of abnormal
pixels is large, increasing the predetermined distance allows an
area for calculating the light amount correction value to be so set
that a large number of normal pixels are present in the area,
whereby the amount of light at the abnormal pixel can precisely be
corrected to a normal amount of light. Further, in this case, when
the number of abnormal pixels is small, reducing the predetermined
distance allows the light amount correction value to be calculated
based on normal pixels closer to the detected abnormal pixel,
whereby a precise light amount correction value can be
calculated.
Further, for example, the predetermined distance may be so set
after edge detection or any other image processing that no edge
portion is contained. In this case, the amounts of light at normal
pixels for the light amount correction value calculation include no
amount of light in an edge portion where the amount of light
greatly changes, whereby a precise light amount correction value
can be calculated.
In the camera according to the aspect of the invention, it is
preferable that the spectroscopic device is capable of changing the
wavelength to be selected.
In the aspect of the invention with the configuration described
above, since the wavelength of light to be separated by the
spectroscopic device can be changed, spectroscopic images
corresponding to a plurality of wavelengths can be acquired.
Replacing the amount of light at an abnormal pixel corresponding to
a specular reflection portion in each of the spectroscopic images
allows acquisition of a high-precision spectroscopic image at each
of the wavelengths.
In the camera according to the aspect of the invention, it is
preferable that the spectroscopic device is a wavelength tunable
Fabry-Perot etalon.
In the aspect of the invention with the configuration described
above, the spectroscopic device is a wavelength tunable Fabry-Perot
etalon. A wavelength tunable Fabry-Perot etalon has a simple
configuration in which a pair of reflection films are simply so
disposed that they face each other and can readily change the
wavelength of light to be separated by changing the dimension of
the gap between the reflection films. Using a thus configured
wavelength tunable Fabry-Perot etalon allows reduction in the size
and thickness of the spectroscopic camera as compared with a case
where an AOTF (acousto-optic tunable filter), an LCTF (liquid
crystal tunable filter), or any other large spectroscopic device is
used.
Another aspect of the invention is directed to an image processing
method in a camera including a light source section that radiates
light toward an object being imaged and an imaging section that
captures light reflected off the object being imaged to acquire an
image, the method including detecting an abnormal pixel in the
image which is a pixel where the ratio of the amount of light at
the pixel to a reference amount of light obtained when a reference
object is irradiated with the light is greater than or equal to a
predetermined value and detecting normal pixels in the image each
of which is a pixel where the ratio is smaller than the
predetermined value and correcting a light amount by calculating a
light amount correction value for correcting the abnormal pixel
based on the amounts of light at normal pixels out of the normal
pixels that are located in a predetermined pixel area around the
abnormal pixel in the image and replacing the amount of light at
the abnormal pixel with the light amount correction value.
In the aspect of the invention, in the pixel detection step, an
abnormal pixel and normal pixels in the captured image are
detected, and in the light amount correction step, a light amount
correction value is calculated based on the amounts of normal
pixels around the abnormal pixel and the amount of light at the
abnormal pixel is replaced with the light amount correction value.
The amount of light at the abnormal pixel can thus be replaced with
an appropriate light amount value, whereby an image having no pixel
showing an abnormal amount of light corresponding to a specular
reflection portion can be acquired, as in the aspect of the
invention described above.
Still another aspect of the invention is directed to a camera that
captures an image produced when an object being imaged is
irradiated with light, and when the amount of light received at a
pixel in the image is an abnormal value, the amount of light at the
pixel is replaced with a light amount correction value calculated
based on the amount of light received at a pixel that is located
within a predetermined distance range from the pixel showing the
abnormal value and shows a normal amount of received light.
In the aspect of the invention, the amount of light at a pixel
showing an abnormal amount of received light is replaced with a
light amount correction value based on the amount of light at a
pixel located around the pixel and showing a normal amount of
received light. As a result, an image having no pixel showing an
abnormal value can be acquired, as in the aspect of the invention
described above.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will be described with reference to the accompanying
drawings, wherein like numbers reference like elements.
FIG. 1 shows a schematic configuration of a spectroscopic analysis
apparatus of a first embodiment according to the invention.
FIG. 2 is a block diagram showing the schematic configuration of
the spectroscopic analysis apparatus of the first embodiment.
FIG. 3 is a plan view showing a schematic configuration of a
wavelength tunable interference filter of the first embodiment.
FIG. 4 is a cross-sectional view of the wavelength tunable
interference filter taken along the line IV-IV in FIG. 3.
FIG. 5 is a flowchart showing a spectroscopic image acquisition
process in the spectroscopic analysis apparatus of the first
embodiment.
FIG. 6 is the flowchart showing the spectroscopic image acquisition
process in the spectroscopic analysis apparatus of the first
embodiment.
FIG. 7 is a flowchart showing a light amount correction process in
the spectroscopic analysis apparatus of the first embodiment.
FIG. 8 shows an example of a spectroscopic image acquired in the
first embodiment.
FIG. 9 shows an example of a spectroscopic image having undergone
light amount correction in the first embodiment.
FIG. 10 is a flowchart showing the light amount correction process
in a second embodiment.
FIG. 11 is a flowchart showing the light amount correction process
in a third embodiment.
DESCRIPTION OF EXEMPLARY EMBODIMENTS
First Embodiment
A spectroscopic analysis apparatus (camera) of a first embodiment
according to the invention will be described below with reference
to the drawings.
Schematic Configuration of Spectroscopic Analysis Apparatus
FIG. 1 is a schematic view showing a schematic configuration of the
spectroscopic analysis apparatus of the first embodiment. FIG. 2 is
a block diagram showing the schematic configuration of the
spectroscopic analysis apparatus.
A spectroscopic analysis apparatus 10 is a camera according to an
embodiment of the invention and an apparatus that captures
spectroscopic images of an object being imaged at a plurality of
wavelengths, analyzes a spectrum in an infrared wavelength region
(target wavelength region in spectroscopic image) at each pixel
based on the captured spectroscopic images, and analyzes the
composition of the object being imaged based on the analyzed
spectra.
The spectroscopic analysis apparatus 10 of the present embodiment
includes an enclosure 11, an imaging module 12, a display 13, an
operation unit 14 (see FIG. 2), and a control unit 15, as shown in
FIG. 1.
Configuration of Imaging Module
The imaging module 12 includes a light incident section 121
(incident light optical system), a light source section 122, a
wavelength tunable interference filter 5 (spectroscopic device), an
imaging section 123, which receives incident light, and a control
substrate 124.
Configuration of Light Incident Section
The light incident section 121 is formed of a plurality of lenses,
as shown in FIG. 1. The light incident section 121 has an angular
field of view limited by the plurality of lenses to a predetermined
angle or smaller and focuses an image of an object under inspection
within the angular field of view onto the imaging section 123. Some
of the plurality of lenses have inter-lens distances that can be
adjusted, for example, by a user who operates the operation unit 14
to enlarge or reduce an acquired image. In the present embodiment,
the lenses that form the light incident section 121 preferably form
a telecentric lens unit. The telecentric lens unit can align the
principal ray of incident light with the direction parallel to the
optical axis, whereby the aligned light can be incident on a fixed
reflection film 54 and a movable reflection film 55 of the
wavelength tunable interference filter 5, which will be described
later, at right angles. Further, when a telecentric lens unit is
used as the lenses that form the light incident section 121, an
aperture is provided in the focal position of the telecentric lens
unit. The aperture has an aperture diameter controlled by the
control unit 15 and can therefore control the angle of incidence of
light incident on the wavelength tunable interference filter 5. The
angle of incidence of the incident light, which is limited by the
group of lenses, the aperture, and other components, is preferably
limited to 20 degrees or smaller with respect to the optical axis,
although the value varies depending on lens design and other
factors.
Configuration of Light Source Section
The light source section 122 radiates light toward an object being
imaged, as shown in FIGS. 1 and 2. The light source section 122 is
formed, for example, of an LED or a laser light source. Using an
LED or a laser light source allows reduction both in size of the
light source section 122 and in power consumption thereof.
Configuration of Wavelength Tunable Interference Filter
FIG. 3 is a plan view showing a schematic configuration of the
wavelength tunable interference filter. FIG. 4 is a cross-sectional
view of the wavelength tunable interference filter taken along the
line IV-IV in FIG. 3.
The wavelength tunable interference filter 5 is a Fabry-Perot
etalon. The wavelength tunable interference filter 5 is, for
example, a rectangular-plate-shaped optical member and includes a
fixed substrate 51, which is formed to a thickness of, for example,
about 500 .mu.m, and a movable substrate 52, which is formed to a
thickness of, for example, about 200 .mu.m. Each of the fixed
substrate 51 and the movable substrate 52 is made, for example, of
soda glass, crystalline glass, quartz glass, lead glass, potassium
glass, borosilicate glass, no-alkali glass, or any of a variety of
other glass materials, or quartz. A first bonding portion 513 of
the fixed substrate 51 and a second bonding portion 523 of the
movable substrate are bonded to each other via a bonding film 53
(first bonding film 531 and second bonding film 532) formed, for
example, of a plasma polymerization film primarily made, for
example, of siloxane so that the fixed substrate 51 and the movable
substrate 52 are integrated with each other.
A fixed reflection film 54 is provided on the fixed substrate 51,
and a movable reflection film 55 is provided on the movable
substrate 52. The fixed reflection film 54 and the movable
reflection film 55 are so disposed that they face each other via a
gap G1. The wavelength tunable interference filter 5 is provided
with an electrostatic actuator 56, which is used to adjust (change)
the dimension of the gap G1. The electrostatic actuator 56 is
formed of a fixed electrode 561 provided on the fixed substrate 51
and a movable electrode 562 provided on the movable substrate 52.
The fixed electrode 561 and the movable electrode 562 face each
other via a gap G2. The fixed electrode 561 and the movable
electrode 562 may be directly provided on substrate surfaces of the
fixed substrate 51 and the movable substrate 52, respectively, or
may be provided thereon via other film members. The dimension of
the G2 is greater than the dimension of the gap G1.
In a filter plan view or FIG. 3 in which the wavelength tunable
interference filter 5 is viewed in the substrate thickness
direction of the fixed substrate 51 (movable substrate 52), a
plan-view center point O of the fixed substrate 51 and the movable
substrate 52 coincides with not only the center point of the fixed
reflection film 54 and the center point of the movable reflection
film 55 but also the center point of a movable portion 521, which
will be described later.
In the following description, a plan view viewed in the substrate
thickness direction of the fixed substrate 51 or the movable
substrate 52, that is, a plan view in which the wavelength tunable
interference filter 5 is viewed in the direction in which the fixed
substrate 51, the bonding film 53, and the movable substrate 52 are
layered on each other is referred to as the filter plan view.
Configuration of Fixed Substrate
The fixed substrate 51 has an electrode placement groove 511 and a
reflection film attachment portion 512 formed therein in an etching
process. The fixed substrate 51 is formed to be thicker than the
movable substrate 52 and is not therefore bent by an electrostatic
attractive force produced when a voltage is applied between the
fixed electrode 561 and the movable electrode 562 or internal
stress induced in the fixed electrode 561.
Further, a cutout 514 is formed at a vertex C1 of the fixed
substrate 51 and exposes a movable electrode pad 564P, which will
be described later and faces the fixed substrate 51 of the
wavelength tunable interference filter 5.
The electrode placement groove 511 is so formed that it has an
annular shape around the plan-view center point O of the fixed
substrate 51 in the filter plan view. The reflection film
attachment portion 512 is so formed that it protrudes from a
central portion of the electrode placement groove 511 in the plan
view described above toward the movable substrate 52. A groove
bottom surface of the electrode placement groove 511 forms an
electrode attachment surface 511A, on which the fixed electrode 561
is disposed. Further, the front end surface of the thus protruding
reflection film attachment portion 512 forms a reflection film
attachment surface 512A.
Further, electrode drawing grooves 511B, which extend from the
electrode placement groove 511 toward the vertices C1 and C2 at the
outer circumferential edge of the fixed substrate 51, are provided
in the fixed substrate 51.
The fixed electrode 561 is disposed on the electrode attachment
surface 511A of the electrode placement groove 511. More
specifically, the fixed electrode 561 is disposed on the electrode
attachment surface 511A in an area facing the movable electrode 562
on the movable portion 521, which will be described later. An
insulating film for ensuring insulation between the fixed electrode
561 and the movable electrode 562 may be layered on the fixed
electrode 561.
A fixed drawn electrode 563 is provided on the fixed substrate 51
and extends from the outer circumferential edge of the fixed
electrode 561 toward the vertex C2. A front end portion of the thus
extending fixed drawn electrode 563 (portion located at vertex C2
of fixed substrate 51) forms a fixed electrode pad 563P, which is
connected to the control substrate 124.
The present embodiment has a configuration in which the single
fixed electrode 561 is provided on the electrode attachment surface
511A but may instead have, for example, a configuration in which
two concentric electrodes formed around the plan-view center point
O are provided on the electrode attachment surface 511A (dual
electrode configuration).
The reflection film attachment portion 512 is coaxial with the
electrode placement groove 511, has a substantially cylindrical
shape having a diameter smaller than that of the electrode
placement groove 511, and has the reflection film attachment
surface 512A facing the movable substrate 52, as described
above.
The fixed reflection film 54 is disposed on the reflection film
attachment portion 512, as shown in FIG. 4. The fixed reflection
film 54 can be formed, for example, of a metal film made, for
example, of Ag or an alloy film made, for example, of an Ag alloy.
The fixed reflection film 54 may instead be formed of a dielectric
multilayer film, for example, having a high refractive layer made
of TiO.sub.2 and a low refractive layer made of SiO.sub.2. The
fixed reflection film 54 may still instead be a reflection film
formed of a metal film (or alloy film) layered on a dielectric
multilayer film, a reflection film formed of a dielectric
multilayer film layered on a metal film (or alloy film), or a
reflection film that is a laminate of a single-layer refractive
layer (made, for example, of TiO.sub.2 or SiO.sub.2) and a metal
film (or alloy film).
An antireflection film may be formed on a light incident surface of
the fixed substrate 51 (surface on which fixed reflection film 54
is not provided) in a position corresponding to the fixed
reflection film 54. The antireflection film can be formed by
alternately layering a low refractive index film and a high
refractive index film on each other, and the thus formed
antireflection film decreases visible light reflectance of the
surface of the fixed substrate 51 whereas increasing visible light
transmittance thereof.
Part of the surface of the fixed substrate 51 that faces the
movable substrate 52, specifically, the surface where the electrode
placement groove 511, the reflection film attachment portion 512,
or the electrode drawing grooves 511B are not formed in the etching
process forms the first bonding portion 513. A first bonding film
531 is provided on the first bonding portion 513 and bonded to a
second bonding film 532 provided on the movable substrate 52,
whereby the fixed substrate 51 and the movable substrate 52 are
bonded to each other as described above.
Configuration of Movable Substrate
The movable substrate 52 has the movable portion 521, which is
circular and formed around the plan-view center point O, a holding
portion 522, which is coaxial with the movable portion 521 and
holds the movable portion 521, and a substrate outer
circumferential portion 525, which is provided in an area outside
the holding portion 522, in the filter plan view or FIG. 3.
Further, the movable substrate 52 has a cutout 524 formed in
correspondence with the vertex C2, and the cutout 524 exposes the
fixed electrode pad 563P when the wavelength tunable interference
filter 5 is viewed from the side where the movable substrate 52 is
present, as shown in FIG. 3.
The movable portion 521 is formed to be thicker than the holding
portion 522. In the present embodiment, for example, the movable
portion 521 is formed to be as thick as the movable substrate 52.
The movable portion 521 is so formed that it has a diameter greater
than at least the diameter of the outer circumferential edge of the
reflection film attachment surface 512A in the filter plan view.
The movable electrode 562 and the movable reflection film 55 are
disposed on the movable portion 521.
An antireflection film may be formed on the surface of the movable
portion 521 that faces away from the fixed substrate 51, as in the
case of the fixed substrate 51. The antireflection film can be
formed by alternately layering a low refractive index film and a
high refractive index film on each other, and the thus formed
antireflection film decreases visible light reflectance of the
surface of the movable substrate 52 whereas increasing visible
light transmittance thereof.
The movable electrode 562 faces the fixed electrode 561 via the gap
G2 and is so formed that it has an annular shape that conforms to
the shape of the fixed electrode 561. A movable drawn electrode 564
is provided on the movable substrate 52 and extends from the outer
circumferential edge of the movable electrode 562 toward a vertex
C1 of the movable substrate 52. A front end portion of the thus
extending movable drawn electrode 564 (portion located at vertex C1
of movable substrate 52) forms the movable electrode pad 564P,
which is connected to the control substrate 124.
The movable reflection film 55 is so disposed on a central portion
of a movable surface 521A of the movable portion 521 that the
movable reflection film 55 faces the fixed reflection film 54 via
the gap G1. The movable reflection film 55 has the same
configuration as that of the fixed reflection film 54 described
above.
In the present embodiment, the dimension of the G2 is greater than
the dimension of the gap G1 as described above by way of example,
but the dimensions of the gaps are not necessarily set this way.
For example, when light under measurement is infrared light or far
infrared light, the dimension of the gap G1 may be greater than the
dimension of the gap G2 depending on the wavelength region of the
light under measurement.
The holding portion 522 is a diaphragm that surrounds the movable
portion 521 and is formed to be thinner than the movable portion
521. The thus configured holding portion 522 is more readily bent
than the movable portion 521 and can therefore displace the movable
portion 521 toward the fixed substrate 51 under a small amount of
electrostatic attractive force. Since the movable portion 521 is
thicker and therefore more rigid than the holding portion 522, the
movable portion 521 is not deformed when the holding portion 522 is
attracted toward the fixed substrate 51 under an electrostatic
attractive force. The movable reflection film 55 disposed on the
movable portion 521 will therefore not be bent, whereby the fixed
reflection film 54 and the movable reflection film 55 can be
consistently maintained parallel to each other.
In the present embodiment, the diaphragm-shaped holding portion 522
is presented by way of example, but the holding portion 522 is not
necessarily formed of a diaphragm. For example, beam-shaped holding
portions disposed at equal angular intervals may be provided around
the plan-view center point O.
The substrate outer circumferential portion 525 is disposed in an
area outside the holding portion 522 in the filter plan view, as
described above. The second bonding portion 523, which faces the
first bonding portion 513, is provided on the surface of the
substrate outer circumferential portion 525 that faces the fixed
substrate 51. The second bonding film 532 is provided on the second
bonding portion 523 and bonded to the first bonding film 513,
whereby the fixed substrate 51 and the movable substrate 52 are
bonded to each other as described above.
Configuration of Imaging Section
The imaging section 123 can, for example, be a CCD, a CMOS device,
or any other image sensor. The imaging section 123 has a
photoelectric element corresponding to each pixel and outputs a
spectroscopic image (image signal) formed of pixels showing the
amounts of light received with the respective photoelectric
elements to the control unit 15.
Configuration of Control Substrate
The control substrate 124 is a circuit substrate that controls the
action of the imaging module 12 and is connected to the light
incident section 121, the light source section 122, the wavelength
tunable interference filter 5, the imaging section 123, and other
components. The control substrate 124 controls the action of each
of the components connected thereto based on a control signal
inputted from the control unit 15. For example, when the user
performs zooming operation, the control substrate 124 moves a
predetermined lens in the light incident section 121 or changes the
aperture diameter of the aperture. Further, when the user performs
operation that triggers capture of a spectroscopic image of an
object being imaged for composition analysis, the control substrate
124 turns on and off the light source section 122 based on a
control signal from the control unit 15. Further, the control
substrate 124 applies a predetermined voltage based on a control
signal from the control unit 15 to the electrostatic actuator 56 in
the wavelength tunable interference filter 5 and outputs a
spectroscopic image captured by the imaging section 123 to the
control unit 15.
Configuration of Display
The display 13 is so provided that it faces a display window of the
enclosure 11. The display 13 may be any display capable of
displaying an image, for example, a liquid crystal panel or an
organic EL panel.
The display 13 in the present embodiment also serves as a tough
panel and hence functions as another operation unit 14.
Configuration of Operation Unit
The operation unit 14 is formed, for example, of a shutter button
provided on the enclosure 11 and a tough panel provided on the
display 13, as described above. When the user performs input
operation, the operation unit 14 outputs an operation signal
according to the input operation to the control unit 15. The
operation unit 14 does not necessarily have the configuration
described above but may have a configuration in which a plurality
of operation buttons or any other components are provided in place
of the touch panel.
Configuration of Control Unit
The control unit 15 is, for example, a combination of a CPU, a
memory, and other components and controls the overall action of the
spectroscopic analysis apparatus 10. The control unit 15 includes a
storage section 16 and a computation section 17, as shown in FIG.
2.
The storage section 16 stores an OS for controlling the overall
action of the spectroscopic analysis apparatus 10, a program for
achieving a variety of functions, and a variety of data. The
storage section 16 has a temporal storage area that temporarily
stores acquired spectroscopic images, composition analysis results,
and other types of information.
An example of the variety of data stored in the storage section 16
is V-.lamda. data representing a drive voltage applied to the
electrostatic actuator 56 in the wavelength tunable interference
filter 5 versus the wavelength of light allowed to pass through the
wavelength tunable interference filter 5 when the drive voltage is
applied.
The storage section 16 further stores correlation data (analytical
curve, for example) representing correlation between a
characteristic quantity extracted from an absorption spectrum
associated with each component of an object under analysis
(absorbance at specific wavelength) and the content of the
component.
The computation section 17 reads the program stored in the storage
section 16 to perform a variety of processes, and the computation
section 17 then functions as a light source controller 171, a
module controller 172, a pixel detector 173 (pixel detection
section), a light amount corrector 174 (light amount correction
section), a composition analyzer 175, and a display controller
176.
The light source controller 171 switches drive operation of the
light source section 122.
The module controller 172 refers to the V-.lamda. data and controls
the electrostatic actuator 56 to change the wavelength of light
allowed to pass through the wavelength tunable interference filter
5 based on the read data. The module controller 172 further
controls the imaging section 123 to cause it to capture
spectroscopic images.
The pixel detector 173 detects an abnormal pixel and a normal pixel
that is not an abnormal pixel based on the amount of light at each
pixel in each acquired spectroscopic image. An abnormal pixel
corresponds to a portion where the light from the light source
section 122 is specularly reflected off the surface of the object
being imaged.
The light amount corrector 174 corrects the amount of light at an
abnormal pixel in each spectroscopic image.
The composition analyzer 175 calculates an optical spectrum at each
pixel based on the spectroscopic images in each of which the amount
of light at an abnormal pixel has been corrected. The composition
analyzer 175 further analyzes the composition of the object being
imaged based on the calculated optical spectrum at each pixel and
the correlation data stored in the storage section 16.
The display controller 176 operates when the module controller 172
controls the imaging module 12 to acquire a captured image and
displays the acquired captured image on the display 13. The display
controller 176 further displays the composition analysis result
provided by the composition analyzer 175 on the display 13.
Specific processes carried out by the computation section 17 will
be described later.
Action of Spectroscopic Analysis Apparatus
The action of the spectroscopic analysis apparatus 10 described
above will next be described below with reference to the
drawings.
To perform composition analysis by using the spectroscopic analysis
apparatus 10 of the present embodiment, an initial process of
acquiring a reference amount of received light for absorbance
calculation is first carried out. The initial process is, for
example, carried out by capturing an image of a reference
calibration plate (reference object) having a perfect diffuse
reflection surface made, for example, of MgO.sub.2 and measuring
the amount of received light (reference amount of light) I.sub.0 at
each wavelength. Specifically, the computation section 17 uses the
module controller 172 to successively change the voltage applied to
the electrostatic actuator 56 to change the wavelength of the
transmitted light, for example, at 10-nm intervals over a
predetermined near-infrared wavelength region (700 to 1500 nm, for
example). The amount of received light at each of the wavelengths
is detected by the imaging section 123 and stored in the storage
section 16.
In this process, the computation section 17 may use the amount of
received light only at one point on the reference calibration plate
as the reference amount of light or may specify a pixel range in
each of the spectroscopic images of the reference calibration
plate, average the amounts of received light at a predetermined
number of pixels or all the pixels in the specified pixel range,
and use the average as the reference amount of light.
A description will next be made of a spectroscopic image
acquisition process (image processing method) in the entire
spectroscopic analysis process using the spectroscopic analysis
apparatus 10. In the spectroscopic analysis apparatus 10 of the
present embodiment, spectroscopic images to be analyzed are
acquired, for example, at 10-nm wavelength intervals over the
infrared region, and the composition analyzer 175 analyzes the
optical spectrum at each pixel in the spectroscopic images to be
analyzed and analyzes an absorption spectrum corresponding to a
component based on the analyzed optical spectrum to determine the
content and other factors of the component contained in the object
being imaged. A description will be made of a process of acquiring
a spectroscopic image to be analyzed (spectroscopic image
processing method) that is carried out before the composition
analysis.
FIGS. 5 to 6 are flowcharts of the spectroscopic image acquisition
process carried out by the spectroscopic analysis apparatus 10.
FIG. 7 is a flowchart of the light amount correction process.
In the spectroscopic image acquisition process, the light source
controller 171 first controls the light source section 122 to cause
it to radiate light toward an object being imaged (step S1), as
shown in FIG. 5. The module controller 172 refers to the V-.lamda.
data stored in the storage section 16, reads a drive voltage
corresponding to a target wavelength, and outputs a control signal
to the control substrate 124 to cause it to apply the drive voltage
to the electrostatic actuator 56 (step S2). The dimension of the
gap between the reflection films 54 and 55 of the wavelength
tunable interference filter 5 is thus changed, and the wavelength
tunable interference filter 5 is ready to transmit light of the
target wavelength.
In steps S1 and S2, light reflected off the object being imaged is
incident through the light incident section 121 to the wavelength
tunable interference filter 5, which transmits light of a
predetermined wavelength according to the dimension of the gap G1
between the reflection films 54 and 55 toward the imaging section
123. The imaging section 123 receives the transmitted light to
capture a spectroscopic image P.sub..lamda.k (step S3). The
captured spectroscopic image P.sub..lamda.k is outputted to the
control unit 15 and stored in the storage section 16.
In the following description, the captured spectroscopic image
P.sub..lamda.k has an image size x.sub.--max.times.y.sub.--max, and
a pixel (x, y) of the spectroscopic image P.sub..lamda.k shows an
amount of light d(x, y).
FIG. 8 shows an example of the acquired spectroscopic image.
Part of the light from the light source section 122 is specularly
reflected off part of the surface of the object being imaged and
incident on the light incident section 121. The spectroscopic image
therefore has a pixel that shows an amount of light (brightness)
greater than the reference amount of light I.sub.0, as shown in
FIG. 8.
The module controller 172 then determines whether or not any
spectroscopic image that has not been acquired is left (step S4).
When it is determined in step S4 that a spectroscopic image that
has not been acquired is left, the control returns to step S2 and
the spectroscopic image acquisition process is continued. The
target wavelength at which a spectroscopic image is acquired
(wavelength corresponding to drive voltage set in step S2) may be
set, for example, in accordance with a component to be analyzed by
the spectroscopic analysis apparatus 10 or may be set by an
operator who performs the measurement as appropriate. For example,
to detect the amounts and calories of lipid, glucide, protein, and
water contained in a food product by using the spectroscopic
analysis apparatus, a wavelength at which the characteristic
quantity of each of at least the lipid, glucide, protein, and water
is obtained may be set as the target wavelength, and it may be
determined in step S4 whether or not a spectroscopic image of each
of the target wavelengths has been acquired.
Spectroscopic images may instead be successively acquired at
predetermined wavelength intervals (10-nm intervals, for
example).
Spectroscopic images P.sub..lamda.k corresponding to target
wavelengths .lamda.k (k=1, 2, 3, . . . ) are thus acquired.
When it is determined in step S4 that the spectroscopic images
P.sub..lamda.k of all the target wavelengths have been acquired, a
process of correcting an abnormal pixel in the spectroscopic images
is carried out.
In this process, the pixel detector 173 first initializes a setting
variable k for selecting a spectroscopic image (k=1) (step S5).
The pixel detector 173 then selects a spectroscopic image
P.sub..lamda.k (step S6) and initializes setting variables i and j
for setting a pixel position on the object under detection (i=1,
j=1) (step S7).
The pixel detector 173 then calculates the ratio of the amount of
light d(i, j) at a pixel (i, j) in the spectroscopic image
P.sub..lamda.k to the reference amount of light I.sub.0
(reflectance ratio) and determines whether or not the reflection
ratio is smaller than or equal to 1 (step S8). That is, it is
determined whether or not the amount of light d(i, j) is smaller
than or equal to the reference amount of light I.sub.0. To
calculate the reflectance ratio for a first spectroscopic image of
a wavelength A, the reference amount of light I.sub.0 at the
wavelength A is used.
When d(i, j)/I.sub.o.ltoreq.1 in step S8, it is determined that the
pixel (i, j) is a "normal pixel," which is not a pixel that
corresponds to a specular reflection portion, and "1" is inputted
to flag data f(i, j) for the pixel (i, j) (step S9).
On the other hand, when d(i, j)/I.sub.o>1 in step S8, it is
determined that the pixel (i, j) is an "abnormal pixel," which is a
pixel that corresponds to a specular reflection portion, and "0" is
inputted to the flag data f(i, j) for the pixel (i, j) (step
S10).
After step S9 or S10, the pixel detector 173 adds "1" to the
setting variable i (step S11: i=i+1) and determines whether or not
the x coordinate specified by the setting variable i in the image
falls within the image size (i.ltoreq.x.sub.--max) (step S12).
When the determination result in step S12 is "Yes", the control
returns to step S8.
On the other hand, when the determination result in step S12 is
"No," the pixel detector 173 initializes the setting variable i
(i=1) and adds "1" to the setting variable j (step S13: j=j+1) and
determines whether or not the y coordinate specified by the setting
variable j in the image falls within the image size
(j.ltoreq.y.sub.--max) (step S14).
When the determination result in step S14 is "Yes," the control
returns to step S8.
On the other hand, when the determination result in step S14 is
"No," the light amount corrector 174 initializes the setting
variables i and j (i=1, j=1) (step S15) as in step S7, as shown in
FIG. 6. The light amount corrector 174 then determines whether or
not the flag data f(i, j) for the pixel (i, j) is "1" (step S16).
That is, the light amount corrector 174 determines whether or not
the pixel (i, j) is a normal pixel.
When the determination result in step S16 is "No," that is, when
the pixel (i, j) is an abnormal pixel, a light amount correction
process (step S100) is carried out.
In the light amount correction process in step S100, the process
shown in FIG. 7 is carried out.
The light amount corrector 174 first acquires a pixel area
(i_min.ltoreq.i.ltoreq.i_max, j_min.ltoreq.j.ltoreq.j_max) formed
of pixels around the abnormal pixel (i, j) as (step S101).
Specifically, in step S101, the following range around the pixel
(i, j) is set as the pixel area: x=i-m1 to i+m2; and y=j-n1 to
j+n2. That is, i_min=i-m1, i_max=i+m2, j_min=j-n1, and j_max=j+n2.
When i-m1<1, i_min is set at 1, and when j-n1<1, j_min is set
at 1. Further, when i+m2>x_max, i_max is set at x_max, and when
j+n2>y_max, j_max is set at y_max.
The values of m1, m2, n1, and n2 may be preset values or may be
values set by the user. When the values of m1, m2, n1, and n2 can
be set by the user, a value for the light amount correction has
changeable correction precision.
The light amount corrector 174 then initializes a normal light
amount sum. "D_sum" and the number of normal pixels "c" (D_sum=0,
c=0) (step S102) and sets the setting variables i and j at the
initial values (i=i_min, j=j_min) (step S103).
The light amount corrector 174 then substitutes D_sum+d(i,
j).times.f(i, j) for the normal light amount sum D_sum and
substitutes c+f(i, j) for the number of normal pixels c (step
S104). When the pixel (i, j) is an abnormal pixel, f(i, j) is "0"
and no addition is therefore made. Only when the pixel (i, j) is a
normal pixel, the amount of light at the pixel (i, j) is added to
the normal light amount sum D_sum, and the number of normal pixels
c is incremented by 1.
The light amount corrector 174 then adds "1" to the setting
variable i (step S105: i=i+1) and determines whether or not the x
coordinate specified by the setting variable i in the image falls
within the pixel area (i.ltoreq.i_max) (step S106).
When the determination result in step S106 is "Yes," the control
returns to step S104.
On the other hand, when the determination result in step S106 is
"No," the light amount corrector 174 sets the setting variable i
back to the initial value (i=i_min), adds "1" to the setting
variable j (step S107: j=j+1), and determines whether or not the y
coordinate specified by the setting variable j in the image falls
within the pixel area (j.ltoreq.j_max) (step S108).
When the determination result in step S108 is "Yes," the control
returns to step S104.
On the other hand, the determination result in step S108 is "No,"
the amount of light d(i, j) at the abnormal pixel (i, j) is
replaced with D_sum/c (step S109). That is, the amount of light at
the abnormal pixel (i, j) is replaced with the average of the
amounts of light at the normal pixels in the pixel area having been
set.
FIG. 9 shows an example of a spectroscopic image having undergone
the light amount correction. The light amount correction process in
step S100 described above replaces an abnormal pixel (i, j)
corresponding to a specular reflection portion, such as the portion
shown in FIG. 8, with a pixel showing a normal light amount value,
such as that shown in FIG. 9.
Referring back to FIG. 6, after step S100 described above is
executed or when the determination result in step S16 is "Yes," the
light amount corrector 174 adds "1" to the setting variable i (step
S17: i=i+1) and determines whether or not the x coordinate
specified by the setting variable i in the image falls within the
image size (i.ltoreq.x_max) (step S18).
When the determination result in step S18 is "Yes," the control
returns to step S16.
On the other hand, when the determination result in step S18 is
"No," the light amount corrector 174 sets the setting variable i
back to the initial value (i=1), adds "1" to the setting variable j
(step S19: j=j+1), and determines whether or not the y coordinate
specified by the setting variable j in the image falls within the
image size (j.ltoreq.y_max) (step S20).
When the determination result in step S20 is "Yes," the control
returns to step S16.
When the determination result in step S20 is "No," the
spectroscopic image P.sub..lamda.k in which the amount of light at
each abnormal pixel is replaced with a light amount correction
value is acquired and stored in the storage section 16.
Thereafter, "1" is added to the setting variable k, which is used
to select a spectroscopic image, (step S21: k=k+1), and it is
determined whether or not k.ltoreq.Kmax (step S22). When the
determination result in step S22 is "Yes," the control returns to
step S6. On the other hand, when the determination result in step
S22 is "No," which means that an spectroscopic image in which the
amount of light at each abnormal pixel has been corrected has been
acquired for each of the target wavelengths .lamda.k (k=1, 2, 3 . .
. Kmax), the spectroscopic image acquisition process is
terminated.
Advantageous Effects of First Embodiment
In the spectroscopic analysis apparatus 10 of the present
embodiment, the light source section 122 radiates light toward an
object being imaged, the wavelength tunable interference filter 5
receives light reflected off the object being imaged and transmits
light of a wavelength according to the dimension of the gap G1
between the reflection films 54 and 55, and the imaging section 123
captures the transmitted light to acquire a spectroscopic image.
The pixel detector 173 calculates the reflectance ratio of the
amount of light at each pixel in the captured spectroscopic image
to the reference amount of light and detects an abnormal pixel
showing a reflectance ratio greater than 1 and a normal pixel
showing a reflectance ratio smaller than or equal to 1. The light
amount corrector 174 then sets a pixel area having a predetermined
range including an abnormal pixel, calculates a light amount
correction value based on the amounts of light of normal pixels
within the pixel area, and replaces the amount of light at the
abnormal pixel with the light amount correction value.
Therefore, even when the light from the light source section 122 is
specularly reflected off the surface of the object being imaged and
hence the resultant spectroscopic image shows abnormal brightness,
the amount of light at the abnormal pixel can be replaced with an
appropriate light amount value based on the amounts of light at the
normal pixels, whereby a spectroscopic image in which no abnormal
pixel is present can be acquired.
Therefore, when spectroscopic measurement is made based on the
acquired spectroscopic image, an optical spectrum at each pixel in
spectroscopic images at a plurality of wavelengths can be precisely
calculated because no abnormal pixel showing a light amount value
higher than the reference amount of light is present in the
spectroscopic images, whereby the composition of the object being
imaged can be precisely analyzed.
In the present embodiment, the light amount corrector 174
calculates a light amount correction value in the form of the
average of the amounts of light at normal pixels within the pixel
area. Therefore, even when the amount of light at an abnormal pixel
is unknown, the amount of light at the abnormal pixel is replaced
with the average of the amounts of light at pixels around the
abnormal pixels so that the amount of light at the abnormal pixel
can be replaced with a value close to an actual amount of light,
whereby a high-precision spectroscopic image can be acquired.
Further, since the amount of light at each normal pixel is
acquired, the average of the amounts of light at the normal pixels
can be readily acquired, whereby the correction process can be
quickly carried out.
The light amount corrector 174, when it sets a pixel area, can
specify the range of the pixel area based on user's input.
Therefore, for example, when a large number of abnormal pixels
present in the pixel area prevent calculation of a precise light
amount correction value, the range of the pixel area can be
widened. On the other hand, narrowing the range of the pixel area
suppresses variation in the light amount values at normal pixels,
whereby more precise light amount correction value can be
calculated. Further, the size of the pixel area along the X
direction (m1, m2) and the size of the pixel area along the Y
direction (n1, n2) can be set with respect to an abnormal pixel (i,
j). Therefore, for example, in a case where an abnormal pixel is
located in the vicinity of an edge portion where the brightness
greatly changes, the pixel area can be so set that it does not
include the edge portion.
Second Embodiment
A second embodiment according to the invention will next be
described with reference to the drawings.
In the first embodiment described above, the amount of light at an
abnormal pixel (i, j) in a spectroscopic image P.sub..lamda.k is
replaced, by way of example, with a light amount correction value
in the form of the average of the amounts of light at normal pixels
in a predetermined pixel area including the abnormal pixel. The
second embodiment differs from the first embodiment described above
in that the light amount correction value is calculated in the form
of the median of the amounts of light at normal pixels.
FIG. 10 is a flowchart showing the light amount correction process
in the second embodiment.
In the following description, the same configurations as those in
the first embodiment and the items having already been described in
the first embodiment have the same reference characters and will
not be described or described in a simplified manner.
In the present embodiment, in the light amount correction process
in step S100, the process in step S101 is carried out to set a
predetermined pixel area including an abnormal pixel (i, j)
detected in a spectroscopic image P.sub..lamda.k as shown in FIG.
10, as in the first embodiment.
Thereafter, the number of normal pixels "c" is initialized or c=0
is inputted (step S111). Further, step S103 is carried out to set
the setting variables i and j at the initial values (i=i_min,
j=j_min).
The light amount corrector 174 then inputs the amount of light d(i,
j) at a normal pixel to a normal value sequence d_n(c) (step S112).
That is, d(i, j).times.f(i, j) is inputted to d_n(c).
Thereafter, f(i, j) is added to the number of normal pixels "c" or
addition of c+f(i, j) is made (step S113). Therefore, when the
pixel (i, j) is a normal pixel, "1" is added to the number of
normal pixels c. The processes in steps S105 to S108 are then
carried out. In the present embodiment, when the determination
result in step S106 is "Yes," the control returns to step S112, and
so does in step S108.
As a result, values are sequentially inputted to the normal value
sequence as follows: d_n(0)=d(i_min, j_min), d_n(1)=d(i_min+1,
j_min) . . . d_n(c)=d(i, j).
When the determination result in step S108 is "No," the light
amount corrector 174 sorts the acquired normal value sequence
d_n(0) to d_n(c-1) in ascending or descending order to acquire a
sorted sequence S_n(1) to S_n(c), which are produced by, (step
S114). For example, when c=5, d_n(0), d_n(1), d_n(2), d_n(3), and
d_n(4) are acquired, which are sorted in ascending order to acquire
S_n(1), S_n(2), S_n(3), S_n(4), and S_n(5).
The light amount corrector 174 then determines whether or not the
number of normal pixels c is an odd number (step S115). When c is
an odd number, the light amount corrector 174 replaces the amount
of light d(i, j) at the abnormal pixel (i, j) with S_n{(c+1)/2}
(step S116).
When the determination result in step S115 is "No" (when it is
determined that c is an even number), the light amount corrector
replaces the amount of light d(i, j) at the abnormal pixel (i, j)
with [S_n{(c/2)+1}+S_n(c/2)]/2 (step S117).
That is, in the present embodiment, the amount of light at the
abnormal pixel (i, j) is replaced with the median of the amounts of
light at normal pixels in a pixel area having been set.
Advantageous Effect of Second Embodiment
In the present embodiment, the light amount corrector 174
calculates a light amount correction value in the form of the
median of the amounts of light at normal pixels in a pixel area and
replaces the amount of light at an abnormal pixel (i, j) with the
light amount correction value.
As a result, even when the amounts of light at normal pixels vary
from each other in the pixel area, selecting a light amount
correction value based on the median in a probability distribution
allows accurate light amount correction and hence acquisition of a
high-precision spectroscopic image.
Third Embodiment
A third embodiment according to the invention will next be
described with reference to the drawings.
In the first and second embodiments described above, the average or
median of the amounts of light at normal pixels is selected as the
light amount correction value in step S100. The present embodiment
differs from the first and second embodiments described above in
that the average of the amounts of light at normal pixels in the
quartile range is acquired.
FIG. 11 is a flowchart showing the light amount correction process
in the third embodiment.
In the present embodiment, the light amount corrector 174 carries
out the processes in steps S101 to S114 as in the second embodiment
to sort the light amount sequence d_n(0) to d_n(c-1), which
represent the amounts of light at normal pixels in a pixel area, to
acquire a sorted sequence S_n(1) to S_n(c), as shown in FIG.
11.
Thereafter, in the present embodiment, the light amount corrector
174 calculates the light amount correction value in the form of the
average of the amounts of light at the normal pixels in the
quartile range as indicated by the following expression and
replaces the amount of light d(i, j) at the abnormal pixel (i, j)
with the light amount correction value (step S121).
.function..times..times..times..times. ##EQU00001##
Advantageous Effect of Third Embodiment
In the present embodiment, the light amount corrector 174
calculates a light amount correction value in the form of the
average of the amounts of light at normal pixels in the quartile
range in a pixel area and replaces the amount of light at an
abnormal pixel (i, j) with the light amount correction value.
Since each of the amounts of light at normal pixels in the quartile
range is a value other than those greater or smaller than the other
values, the amount of light at an abnormal pixel can be corrected
by using a more accurate amount of light, whereby a high-precision
spectroscopic image can be acquired.
Other Embodiments
The invention is not limited to the embodiments described above,
and variations, modifications, and other improvements to the extent
that the advantage of the invention is achieved fall within the
scope of the invention.
For example, in the embodiments described above, the spectroscopic
analysis apparatus 10 is presented by way of example. The invention
is also applicable to a typical spectroscopic camera that is not
intended for composition analysis or other types of analysis of an
object being imaged.
In the first to third embodiments described above, the
spectroscopic analysis apparatus 10 including a spectroscopic
camera that acquires a spectroscopic image is presented by way of
example, but the spectroscopic analysis apparatus 10 does not
necessarily include a spectroscopic camera. For example, the
invention is also applicable to a typical camera that captures, for
example, a color image. In this case as well, based on the amount
of light at each pixel in a captured image (for example, each of R,
G, and B monochromatic images that form a color image captured by
using RGB color filters) and a reference amount of light, the
reflection ratio at the pixel is calculated, the pixel is detected
as an abnormal pixel when the reflection ratio is greater than a
predetermined value whereas detected as a normal pixel when the
reflection ratio is smaller than or equal to the predetermined
value, and the amount of light at the abnormal pixel is corrected
based on the amount of light at the normal pixel.
Further, the above first to third embodiments have been described
with reference to the cases where the amount of light at an
abnormal pixel is corrected by using the average of the amounts of
light at normal pixels, by using the median of the amounts of light
at normal pixels, and by using the average of the amounts of light
at normal pixels in the quartile range, respectively, but the
amount of light at an abnormal pixel is not necessarily corrected
as described above.
For example, the light amount correction value may be calculated
based on the amounts of light at pixels around an abnormal pixel by
using spline interpolation, and the amount of light at the abnormal
pixel may be replaced with the calculated light amount correction
value. In this case, the amount of light at a pixel in a corrected
image smoothly changes to the amount of light at an adjacent pixel,
whereby an inconvenient situation in which the amount of light at
an abnormal pixel is replaced with an unnatural amount of light is
avoided.
The above first to third embodiments have been described with
reference to the case where it is determined whether or not the
reflection ratio (d(i, j)/I.sub.0) is smaller than or equal to a
predetermined value (1) set in advance, but the reflection ratio is
not necessarily used. For example, an input section that receives a
predetermined value inputted by the user who operates the operation
unit 14 and uses the inputted predetermined value to determine
whether a pixel in question is a normal pixel or an abnormal pixel
may be provided.
In this case, in a situation in which correct light amount
correction cannot be made, for example, when the reflection ratios
at the pixels in an acquired image are high as a whole, abnormal
pixel detection sensitivity can be lowered by changing the
predetermined value described above, which serves as a threshold,
as appropriate, whereby an image in which the proportion of
specular reflection portion is reduced can be acquired.
Further, the input section does not necessarily acquire a value
based on user's operation. For example, when the number of abnormal
pixels in an acquired image is greater than a predetermined upper
limit, a process of lowering the predetermined value described
above, which is used to determine whether a pixel in question is a
normal or abnormal pixel, may be carried out. In this case as well,
the same advantageous effect as that described above can be
provided.
In the first to third embodiments described above, the light amount
corrector 174, when it sets a pixel area around an abnormal pixel
(i, j), may carry out a process of notifying a user of an abnormal
situation with an alert (notification displayed on the display,
audio notification, or other forms of notification) when the number
of abnormal pixels contained in the pixel area is greater than or
equal to a predetermined first threshold or when the proportion of
the number of normal pixels to the number of abnormal pixels is
smaller than a second threshold. Instead of the notification, a
spectroscopic image may be reacquired by performing the measurement
(image capturing) again.
Further, the light amount corrector 174 may carry out, when it sets
a pixel area, a process of widening the set range of the pixel area
when the number of abnormal pixels contained in the pixel area is
greater than or equal to the predetermined first threshold or when
the proportion of the number of normal pixels to the number of
abnormal pixels is smaller than the second threshold. That is, in
step S101, each of the values of m1, m2, n1, and n2 may be
increased by a predetermined amount, and a pixel area may be set
again in accordance with the increased values. In this case, the
number of normal pixels is likely to be greater than the number of
abnormal pixels, whereby a light amount correction value can be
calculated based on the larger number of normal pixels.
Further, the light amount corrector 174 may carry out, when it sets
a pixel area, a process of narrowing the set range of the pixel
area when the number of abnormal pixels contained in the pixel area
is smaller than or equal to a predetermined third threshold or when
the proportion of the number of normal pixels to the number of
abnormal pixels is greater than a predetermined fourth threshold.
That is, in step S101, each of the values of m1, m2, n1, and n2 may
be reduced by a predetermined amount, and the pixel area may be set
in accordance with the reduced values for calculation of a light
amount correction value based on normal pixels closer to the
abnormal pixel (i, j). In this case, a more accurate light amount
correction value can be calculated based on normal pixels close to
the abnormal pixel (i, j).
Further, the light amount corrector 174 may determine whether or
not an edge portion where the amount of light greatly changes from
a pixel to an adjacent pixel is present in the pixel area. When
determining that such an edge portion is present, the light amount
corrector 174 may further detect the positional relationship
between an abnormal pixel (i, j) and the pixels in the edge portion
and set a pixel area that does not include the edge portion. In
this case, the edge portion or any other area where the amount of
light greatly changes can be excluded, whereby an accurate light
amount correction value can be calculated.
In each of the embodiments described above, the wavelength tunable
interference filter 5 may be accommodated in a package, and the
packaged wavelength tunable interference filter 5 may be
incorporated in the spectroscopic analysis apparatus 10. In this
case, the package can be exhausted to a vacuum and sealed to
improve the response of the electrostatic actuator 56 in the
wavelength tunable interference filter 5 when the electrostatic
actuator 56 is driven by voltage application.
In each of the embodiments described above, after the reference
calibration plate having a perfect diffuse reflection surface is
irradiated with light, the amount of light received from the
reference calibration plate is used as the reference amount of
light I.sub.0, and the pixel detector 173 calculates the
reflectance ratio based on the reference amount of light I.sub.0.
Instead, for example, a reference calibration plate that absorbs
part of light incident on the surface thereof and hence does not
provide perfect diffuse reflection may be used. In this case, the
pixel detector 173 can determine whether or not a pixel in question
is an abnormal pixel by determining whether or not the reflectance
ratio is smaller than or equal to a predetermined value smaller
than to 1.
The wavelength tunable interference filter 5 is configured to
include the electrostatic actuator 56, which changes the dimension
of the gap between the reflection films 54 and 55 based on voltage
application, but the dimension of the gap is not necessarily
changed this way.
For example, an induction actuator having a first induction coil
provided in place of the fixed electrode 561 and a second induction
coil or a permanent magnet provided in place of the movable
electrode 562 may be used.
Further, a piezoelectric actuator may be used in place of the
electrostatic actuator 56. In this case, for example, a lower
electrode layer, a piezoelectric film, and an upper electrode layer
are layered on each other and disposed at the holding portion 522,
and a voltage applied between the lower electrode layer and the
upper electrode layer can be changed as an input value to expand or
contract the piezoelectric film so as to bend the holding portion
522.
Further, each of the embodiments described above shows the case
where the wavelength tunable interference filter 5 is configured as
a Fabry-Perot etalon and includes the fixed substrate 51 and the
movable substrate 52 so bonded to each other that they face each
other with the fixed reflection film 54 provided on the fixed
substrate 51 and the movable reflection film 55 provided on the
movable substrate 52, but the configuration of the wavelength
tunable interference filter 5 is not limited thereto.
For example, the wavelength tunable interference filter 5 may be so
configured that the fixed substrate 51 and the movable substrate 52
are not bonded to each other but a gap changer that changes the gap
between the reflection films, such as a piezoelectric device, is
provided between the substrates.
Further, the wavelength tunable interference filter 5 is not
necessarily formed of two substrates. For example, a wavelength
tunable interference filter so configured that two reflection films
are layered on a single substrate with a sacrifice layer between
the reflection films and the sacrifice layer is etched away or
otherwise removed to form a gap may be used.
Moreover, as the spectroscopic device, an AOTF (acousto-optic
tunable filter), an LCTF (liquid crystal tunable filter), or any
other similar device may be used. In this case, however, size
reduction of the spectroscopic camera (spectroscopic analysis
apparatus 10) is likely to be difficult. It is therefore preferable
to use a Fabry-Perot etalon.
Further, in each of the embodiments described above, the wavelength
tunable interference filter 5 that can change the wavelength of
transmitted light by changing the gap G1 between the reflection
films 54 and 55 is presented by way of example, but the wavelength
tunable interference filter 5 is not necessarily used. For example,
a wavelength-fixed interference filter (Fabry-Perot etalon) may be
used. In this case, the amount of light at an abnormal pixel can be
appropriately corrected in a spectroscopic image of a specific
wavelength according to the dimension of the gap between the
reflection films of the interference filter.
In addition, the specific structure according to any of the
embodiments of the invention can be changed as appropriate in
actual implementation of the invention to any other structure to
the extent that the advantage of the invention is achieved.
The entire disclosure of Japanese Patent Application No.
2013-050096, filed Mar. 13, 2013 is expressly incorporated by
reference herein.
* * * * *