Uncooled Infrared Imaging Device

Abstract

An uncooled infrared imaging device according to an embodiment includes: reference pixels formed on a semiconductor substrate and arranged in at least one row; and infrared detection pixels arranged in the remaining rows and detecting incident infrared rays. Each of the reference pixels includes a first cell located above a first concave portion. The first cell includes a first thermoelectric conversion unit having a first infrared absorption film; and a first thermoelectric conversion element. Each of the infrared detection pixels includes a second cell located above a second concave portion, and having a larger area than the first cell. The second cell includes: a second thermoelectric converting unit located above the second concave portion; and first and second supporting structure units supporting the second thermoelectric converting unit above the second concave portion. The second thermoelectric converting unit includes: a second infrared absorption film; and a second thermoelectric conversion element.

Description

CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application is based upon and claims the benefit of priority from prior Japanese Patent Application No. 2012-68158 filed on Mar. 23, 2012 in Japan, the entire contents of which are incorporated herein by reference.

FIELD

[0002] An embodiment described herein relates generally to an uncooled infrared imaging device.

BACKGROUND

[0003] Infrared rays are radiated from a heat source even in the dark, and characteristically have higher permeability than visible light in smoke or fog. Accordingly, infrared imaging can be performed during day and night. Also, temperature information about an object can be obtained through infrared imaging. In view of this, infrared imaging can be applied to a wide variety of fields such as the defense field and the fields of security cameras and fire detecting cameras.

[0004] In recent years, "uncooled infrared imaging devices" that do not require cooling mechanisms have been actively developed. An uncooled or thermal infrared imaging device converts an incident infrared ray of approximately 10 .mu.m in wavelength into heat with an absorption mechanism, and further converts the temperature change at the heat sensitive unit caused by the small amount of heat into an electrical signal with a thermoelectric conversion means. The uncooled infrared imaging device reads the electrical signal, to obtain infrared image information.

[0005] For example, there have been infrared sensors each using a silicon pn junction that converts a temperature change into a voltage change by applying a constant forward current. Characteristically, such infrared sensors can be mass-produced through a silicon LSI manufacturing process using a SOI (Silicon On Insulator) substrate as a semiconductor substrate. Also, the rectification properties of the silicon pn junctions serving as the thermoelectric conversion means are utilized to realize the row select function, and accordingly, the pixels can be designed to have very simple structures.

[0006] One of the indicators of the performance of an infrared sensor is NETD (Noise Equivalent Temperature Difference), which indicates the temperature resolution of the infrared sensor. It is critical to reduce the NETD, or reduce the infrared sensor temperature difference equivalent to noise. To do so, it is necessary to increase signal sensitivity and reduce noise.

[0007] Thermoelectric conversion elements are sensitive to temperature components other than temperature rises caused by incident infrared rays, or to the temperature of the semiconductor substrate and the self-heating temperature at the time of flowing of current. To correct those "offset temperatures", reference pixels are provided.

[0008] Like an infrared detection pixel, a conventional reference pixel reflects the influence of the temperature of the semiconductor substrate, but has a different self-heating temperature from that of an infrared detection pixel. The difference in self-heating temperature is much larger than a signal generated from an incident infrared ray, and therefore, it is necessary to correct the difference.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1 is a circuit diagram of an uncooled infrared imaging device according to an embodiment;

[0010] FIGS. 2(a) and 2(b) are a plan view and a cross-sectional view of an infrared detection pixel according to the embodiment, respectively;

[0011] FIGS. 3(a) and 3(b) are a plan view and a cross-sectional view of a reference pixel according to the embodiment, respectively; and

[0012] FIGS. 4(a) and 4(b) are graphs showing the self-heating amounts in an infrared detection pixel and a reference pixel, respectively.

DETAILED DESCRIPTION

[0013] An uncooled infrared imaging device according to an embodiment includes: a semiconductor substrate; an imaging region formed on the semiconductor substrate, a plurality of pixels being arranged in a matrix form in the imaging region, the plurality of pixels including a plurality of reference pixels arranged in at least one row, and a plurality of infrared detection pixels arranged in the remaining rows, each of the infrared detection pixels being configured to detect an incident infrared ray, each of the reference pixels including a first cell located above a first concave portion formed in a surface portion of the semiconductor substrate, the first cell including a first thermoelectric converting unit including a first infrared absorption film and a first thermoelectric conversion element, the first infrared absorption film being configured to absorb the incident infrared ray and convert the incident infrared ray into heat, the first thermoelectric conversion element being configured to convert the heat obtained by the first infrared absorption film into an electrical signal, each of the infrared detection pixels including a second cell located above a second concave portion formed in a surface portion of the semiconductor substrate, the second cell including a second thermoelectric converting unit located above the second concave portion and first and second supporting structure units supporting the second thermoelectric converting unit above the second concave portion, the second thermoelectric converting unit including a second infrared absorption film and a second thermoelectric conversion element, the second infrared absorption film being configured to absorb the incident infrared ray and convert the incident infrared ray into heat, the second thermoelectric conversion element being configured to convert the heat obtained by the second infrared absorption film into an electrical signal, the first cell having a smaller area than the second cell when viewed from an incident direction of the infrared ray; a plurality of row select lines formed in the imaging region, the plurality of row select lines corresponding to the rows of the plurality of pixels, each of the row select lines being connected to one end of the thermoelectric conversion element of each pixel of the corresponding one of the rows, each of the row select lines selecting the pixels of the corresponding one of the rows; and a plurality of signal lines formed in the imaging region, the plurality of signal lines corresponding to columns of the plurality of pixels, each of the signal lines being connected to the other end of the thermoelectric conversion element of each pixel of the corresponding one of the columns, each of the signal lines being used for reading the electrical signal from each pixel of the corresponding one of the columns.

[0014] The following is a description of an embodiment, with reference to the accompanying drawings.

[0015] FIG. 1 shows an uncooled infrared imaging device (hereinafter also referred to as the infrared sensor) according to an embodiment. The infrared sensor of this embodiment includes an imaging region formed on a semiconductor substrate. In this imaging region, there are two reference pixels 11.sub.1 and 11.sub.2 that are arranged in one row and two columns, and infrared detection pixels 12.sub.ij (i, j=1, 2) arranged in two rows and two columns, for example. Although an imaging region normally accommodates more pixels than the above, the imaging region in this embodiment accommodates pixels, or reference pixels and infrared detection pixels, which are arranged in three rows and two columns, for descriptive purposes. Each of the pixels includes a pn junction (diodes) serving as a thermoelectric conversion element, which will be described later in detail.

[0016] The first terminal (the anode-side terminal of the pn junction) of each of the reference pixels 11.sub.1 and 11.sub.2 is connected to a row select line 45.sub.0. The second terminals (the cathode-side terminals of the pn junctions) of the reference pixels 11.sub.1 and 11.sub.2 are connected to vertical signal lines (hereinafter also referred to simply as the signal lines) 44.sub.1 and 44.sub.2, respectively. The respective first terminals (the anode-side terminals of the pn junctions) of the infrared detection pixels 12.sub.11 and 12.sub.12 of the first row are connected to a row select line 45.sub.1. The second terminals (the cathode-side terminals of the pn junctions) of the infrared detection pixels 12.sub.11 and 12.sub.12 are connected to the vertical signal lines 44.sub.1 and 44.sub.2, respectively. The respective first terminals (the anode-side terminals of the pn junctions) of the infrared detection pixels 12.sub.21 and 12.sub.22 of the second row are connected to a row select line 45.sub.2. The second terminals (the cathode-side terminals of the pn junctions) of the infrared detection pixels 12.sub.21 and 12.sub.22 are connected to the vertical signal lines 44.sub.1 and 44.sub.2, respectively.

[0017] Each of the row select lines 45.sub.0, 45.sub.1, and 45.sub.2 is connected to a row select circuit 5. The rows are sequentially selected one by one by the row select circuit 5, and a bias voltage Vd is applied to each selected row. One terminal of each signal line 44.sub.i (i=1, 2) is connected to a power source Vss via a load transistor 41.sub.i, and the other terminal is connected to the positive-side input terminal of a differential amplifier 20.sub.i via a coupling capacitor 22.sub.i. The load transistors 41.sub.1 and 41.sub.2 operate in a saturation region, and supply a constant current to the pixels of a selected row in accordance with a gate voltage GL1 applied to the gates thereof. That is, the load transistors 41 function as constant current sources. The source voltage of each of the load transistors 41.sub.1 and 41.sub.2 is represented by Vd0.

[0018] A voltage V2 is applied to the negative-side input terminal of each differential amplifier 20.sub.i(i=1, 2), and the output terminal of each differential amplifier 20.sub.i is connected to a read line 33 via a read transistor 32.sub.i. Upon receipt of a select signal from a column select circuit 6, each read transistor 32.sub.i(i=1, 2) is switched on.

[0019] A feedback switch 23.sub.i and a feedback capacitor 24.sub.i are connected in parallel between the positive-side input terminal and the output terminal of each differential amplifier 20.sub.i (i=1, 2). A coupling capacitor 22.sub.i (i=1, 2), a differential amplifier 20.sub.i, a feedback switch 23.sub.i, and a feedback capacitor 24.sub.i constitute a differential clamp circuit. As will be described later, this differential clamp circuit holds signals that are output from the reference pixel connected to the corresponding signal line, and amplifies a differential signal when an infrared detection pixel connected to the corresponding signal line is selected.

[0020] When the row select circuit 5 applies the bias voltage Vd to a selected row, a series voltage Vd-Vd0 is applied to the pn junctions of the reference pixels or the infrared detection pixels connected to the selected row. For example, if the row select line 45.sub.0 is selected, the series voltage Vd-Vd0 is applied to the pn junctions of the reference pixels 11.sub.1 and 11.sub.2. If the row select line 45.sub.1 is selected, the series voltage Vd-Vd0 is applied to the pn junctions of the infrared detection pixels 12.sub.11 and 12.sub.12. If the row select line 45.sub.2 is selected, the series voltage Vd-Vd0 is applied to the pn junctions of the infrared detection pixels 12.sub.21 and 12.sub.22. All the pn junctions of the pixels of the unselected rows are inversely-biased. Therefore, the unselected row select lines are separated from the signal lines 44.sub.1 and 44.sub.2. That is, it can be said that the pn junctions each have a pixel select function.

[0021] The potential of each of the vertical signal lines 44.sub.1 and 44.sub.2 when any infrared ray is not received is defined as Vs1. Upon receipt of an infrared ray, the pixel temperature of each infrared detection pixel 12.sub.ij (i, j=1, 2) becomes higher. Accordingly, the potential of each of the vertical signal lines 44.sub.1 and 44.sub.2 becomes higher than Vs1. For example, when the object temperature varies by 1 K (Kelvin), the temperature of an infrared detection pixel 12.sub.ij (i, j=1, 2) varies by approximately 5 mK. With the thermoelectric conversion efficiency being 10 mV/K, the potential of each of the vertical signal lines 44.sub.1 and 44.sub.2 increases by approximately 50 .mu.V, which is much smaller than the increase of the bias voltage Vd.

[0022] Upon receipt of an infrared ray, the potential of each of the vertical signal line 44.sub.1 and 44.sub.2 becomes Vd-(Vf0-Vsig-Vsh). Here, Vf0 represents the forward voltage of a pn junction when no infrared ray is received, and Vsig represents the voltage signal based on the temperature rise in the later described thermoelectric converting unit of an infrared detection pixel. Vsh represents the change in voltage due to the Joule heat generated when a current is applied to the pn junction. The self-heating amount is expressed by the following equation (1):

T cell ( t ) = I f V f G th { 1 - exp ( - t C th / G th ) } ( 1 ) ##EQU00001##

[0023] Here, I.sub.f represents the amount of current determined by the operating points of the load transistors 41.sub.1 and 41.sub.2, V.sub.f represents the forward voltage of the pn junction, t represents the time elapsed from the start of flowing the current, and C.sub.th and G.sub.th represent the heat capacity and the heat conductance of the infrared detection pixel 12.sub.ij (i, j=1, 2), respectively. Those items will be described later. In a case where t is sufficiently smaller than the value expressed by C.sub.th/G.sub.th, the equation (1) is approximated by the following equation (2), and the temperature increases in proportion to time.

T cell ( t .fwdarw. 0 ) = I f V f C th t ( 2 ) ##EQU00002##

[0024] FIG. 2(a) is a plan view of an infrared detection pixel 12.sub.ij (i, j=1, 2). FIG. 2(b) is a cross-sectional view of the infrared detection pixel 12.sub.ij (i, j=1, 2), taken along the section line A-A of FIG. 2(a). The infrared detection pixel 12.sub.ij (i, j=1, 2) includes a cell 160 formed on a SOI substrate. The SOI substrate includes a supporting substrate 181, a buried insulating layer (a BOX layer) 182, and a SOI (Silicon-On-Insulator) layer formed with silicon single crystals. The SOI substrate has a concave portion 185 formed in a surface portion thereof. The concave portion 185 is formed by removing part of the supporting substrate 181.

[0025] The cell 160 includes a thermoelectric converting unit 161, and supporting structure units 167a and 167b that support the thermoelectric converting unit 161 above the concave portion 185. The thermoelectric converting unit 161 includes diodes (two diodes in FIG. 2(a)) 162 connected in series, an interconnect 163 connecting those diodes 162, and an infrared absorption film 164 formed to cover the diodes 162 and the interconnect 163.

[0026] The supporting structure unit 167a includes: a connective interconnect 168a that has one end connected to the corresponding vertical signal line and has the other end connected to one end of the series circuit formed with the series-connected diodes 162; and an insulating film 169a coating the connective interconnect 168a. The other supporting structure unit 167b includes: a connective interconnect 168b that has one end connected to the corresponding row select line and has the other end connected to the other end of the series circuit formed with the series-connected diodes 162; and an insulating film 169b coating the connective interconnect 168b.

[0027] The infrared absorption film 164 generates heat upon receipt of an incident infrared ray. The diodes 162 convert the heat generated by the infrared absorption film 164 into an electrical signal. The supporting structure units 167a and 167b have elongate shapes so as to surround the thermoelectric converting unit 161. With this arrangement, the thermoelectric converting unit 161 is supported on the SOI substrate, while being substantially heat-insulated from the SOI substrate.

[0028] Having such a structure, each infrared detection pixel 12.sub.ij (i, j=1, 2) stores the heat generated from an incident infrared ray, and can output the voltage based on the heat to the signal line 44.sub.1 or 44.sub.2. The bias voltage Vd from the row select line 45.sub.1 or 45.sub.2 is transferred to the thermoelectric converting unit 161 via the interconnect 168b. The signal that has passed the thermoelectric converting unit 161 is transferred to the vertical signal line 44.sub.1 or 44.sub.2 via the interconnect 168a.

[0029] FIG. 3(a) is a plan view of a reference pixel 11.sub.i (i=1, 2). FIG. 3(b) is a cross-sectional view of the reference pixel 11.sub.i (i=1, 2), taken along the section line A-A of FIG. 3(a). The reference pixel 11.sub.i (i=1, 2) includes a cell 160a formed on a SOI substrate. The SOI substrate includes a supporting substrate 181, a buried insulating layer (a BOX layer) 182, and a SOI (Silicon-On-Insulator) layer formed with silicon single crystals. The SOI substrate has a concave portion 185a formed in a surface portion thereof. The concave portion 185a is formed by removing part of the supporting substrate 181.

[0030] The cell 160a includes a thermoelectric converting unit 161a formed on the buried insulating layer 182. The thermoelectric converting unit 161a includes diodes (two diodes in FIG. 3(a)) 162a connected in series, an interconnect 163a connecting those diodes 162a, and an infrared absorption film 164a formed to cover the diodes 162a and the interconnect 163a. One end of the series circuit formed with the series-connected diodes 162a is connected to a vertical signal line via an interconnect 165a, and the other end of the series circuit is connected to a row select line via an interconnect 165b. The interconnects 165a and 165b are formed in regions where the concave portion 185a is not formed. The concave portion 185a is formed below the thermoelectric converting unit 161a. However, the thermoelectric converting unit 161a is connected to the portion of the buried insulating layer 182 formed outside the region where the concave portion 185a is formed and to the portion of the insulating film 164 formed on the portion of the buried insulating layer 182, via the buried insulating layer 182 and the insulating film 164 formed on the buried insulating layer 182. Accordingly, unlike the thermoelectric converting unit 161 of each infrared detection pixel, the thermoelectric converting unit 161a does not need to be supported above the concave portion 185a by elongate supporting structure units. That is, unlike the thermoelectric converting unit 161 of each infrared detection pixel, the thermoelectric converting unit 161a is not heat-insulated by elongate supporting structure units. Therefore, the heat generated from an infrared ray is smaller by several digits than that generated by an infrared detection pixel, and can be ignored. That is, in each reference pixel, the heat conductance G.sub.th, which is an indicator of heat insulating properties, is much higher than that of each infrared detection pixel, and heat easily escapes from each reference pixel.

[0031] The unit of the heat conductance G.sub.th is W/K, indicating how many watts of energy is transferred in a case where a heat conductor exists between two heat baths between which the temperature difference is 1 K. The heat conductance G.sub.th is expressed as G.sub.th=.kappa.S/L(W/K), where .kappa.(W/(Km)) represents the heat conductivity, S (m.sup.2) represents the cross-sectional area, and L (m) represents the length of the material responsible for heat conduction, or of the connective interconnects 168a and 168b and the insulating films 169a and 169b constituting the supporting structure units 167a and 167b. That is, a structure with a larger cross-sectional area and a shorter length has higher heat conductance.

[0032] Meanwhile, the heat capacity C.sub.th is the indicator of how many joules of energy is required to increase the temperature of an object by 1K, and the unit thereof is J/K. The heat capacity C.sub.th is expressed as C.sub.th=cdV, where c (J/kg) represents the specific heat of the material, V (m.sup.3) represents the volume of the material, and d (kg/m.sup.3) represents the density of the material.

[0033] The heat conductance G.sub.th.sub.--.sub.IMG and the heat capacity C.sub.th.sub.--.sub.IMG of each infrared detection pixel 12.sub.ij (i, j=1, 2) are expressed as:

G.sub.th.sub.--.sub.IMG=.kappa..sub.2S/L

C.sub.th.sub.--.sub.IMG=c.sub.2d.sub.2LcWcHc (3)

[0034] Here, S represents the cross-sectional area of the supporting structure units 167a and 167b, L represents the length of the supporting structure units 167a and 167b, Lc and We represent the length and the width of the thermoelectric converting unit 161, respectively, and Hc represents the height of the thermoelectric converting unit 161 inclusive of the thickness of the buried insulating film 182. Meanwhile, .kappa..sub.2, c.sub.2, and d.sub.2 represent the heat conductivity, the specific heat, and the density of each of the supporting structure units 167a and 167b, respectively.

[0035] The heat conductance G.sub.th.sub.--.sub.TB and the heat capacity C.sub.th.sub.--.sub.TB of each reference pixel 11.sub.i (i=1, 2) are expressed as:

G.sub.th.sub.--.sub.TB=.kappa..sub.1S.sub.TB/L.sub.TB

C.sub.th.sub.--TB=c.sub.1d.sub.1LcWc.sub.TBHc (4)

[0036] Here, S.sub.TB and L.sub.TB represent the cross-sectional area and the length of the connecting portion 166 between the thermoelectric converting unit 161a and the interconnects 165a and 165b, respectively, Lc and Wc.sub.TB represent the length and the width of the thermoelectric converting unit 161a, respectively, and Hc represents the height of the thermoelectric converting unit 161a inclusive of the thickness of the buried insulating film 182. Meanwhile, .kappa..sub.1, c.sub.1, and d.sub.1 represent the heat conductivity, the specific heat, and the density of the connecting portion 166 between the thermoelectric converting unit 161a and the interconnects 165a and 165b, respectively.

[0037] In each reference pixel 11.sub.i (i=1, 2), the plane pattern of the concave portion 185a is designed so that the width Wc.sub.TB of the thermoelectric converting unit 161a is smaller than the width Wc of the thermoelectric converting unit 161 of each infrared detection pixel 12.sub.ij (i, j=1, 2).

[0038] Therefore, the heat conductance of each reference pixel is lower than the heat conductance of each infrared detection pixel, and the heat capacity of each reference pixel is smaller than the heat capacity of each infrared detection pixel. That is, the following inequalities are satisfied:

G.sub.th.sub.--.sub.TB<G.sub.th.sub.--.sub.IMG

C.sub.th.sub.--.sub.TB<C.sub.th.sub.--.sub.IMG

[0039] FIGS. 4(a) and 4(b) show graphs of the self-heating amounts in each infrared detection pixel 12.sub.ij (i, j=1, 2) and each reference pixel 11.sub.i (i=1, 2). FIG. 4(a) illustrates the self-heating in a case where the heat capacity C.sub.th.sub.--.sub.TB of each reference pixel 11.sub.i (i=1, 2) is equal to the heat capacity C.sub.th.sub.--.sub.IMG of each infrared detection pixel 12.sub.ij (i, j=1, 2), or where the areas of the cells 160 and 160a are the same. Once flowing of current is started, the temperature of each pixel increases according to the equation (1). As for each infrared detection pixel 12.sub.ij (i, j=1, 2), the heat conductance G.sub.th.sub.--.sub.IMG is very low, and the ratio C.sub.th.sub.--.sub.IMG/G.sub.th.sub.--.sub.IMG is high. Accordingly, the thermal time constant is high, and the temperature increases according to the equation (2) at the time of short-time flowing of current. As for each reference pixel 11.sub.i (i=1, 2), the heat conductance G.sub.th.sub.--.sub.TB is higher than the heat conductance of each infrared detection pixel. Therefore, as indicated by the equation (1), the temperature increase rate becomes lower with time. In this case, the self-heating temperature rises of both pixels differ from each other when the flowing of current ends. This difference is on the order of approximately 100 mK in a case where a current of 1 .mu.A is applied, and is larger than a signal with respect to an object temperature change by about two digits.

[0040] However, in a case where the heat capacity C.sub.th.sub.--.sub.TB of each reference pixel 11.sub.i (i=1, 2) is made smaller than the heat capacity C.sub.th.sub.--.sub.IMG of each infrared detection pixel 12.sub.ij (i, j=1, 2), or where the area of the cell 160a is made smaller than the area of the cell 160 as in this embodiment, the self-heating temperatures can be made the same when the flowing of current ends, as shown in FIG. 4(b). At this point, the design parameters of each pixel should satisfy the following equation:

I f V f G th_IMG { 1 - exp ( - tsel C th_IMG / G th_IMG ) } = I f V f G th_TB { 1 - exp ( - tsel C th_TB / G th_TB ) } ##EQU00003##

[0041] Since each infrared detection pixel 12.sub.ij (i, j=1, 2) and each reference pixel 11.sub.i (i=1, 2) are operated under the same current and voltage conditions, I.sub.f and V.sub.f are equal. Therefore, the following equation should be satisfied:

1 G th_IMG { 1 - exp ( - tsel C th_IMG / G th_IMG ) } = 1 G th_TB { 1 - exp ( - tsel C th_TB / G th_TB ) } ( 5 ) ##EQU00004##

[0042] Here, G.sub.th.sub.--.sub.IMG and G.sub.th.sub.--.sub.TB represent the heat conductance of each infrared detection pixel and the heat conductance of each reference pixel, respectively, C.sub.th.sub.--.sub.IMG and C.sub.th.sub.--.sub.TB represent the heat capacity of the cell of each infrared detection pixel and the heat capacity of the cell of each reference pixel, respectively, tsel represents the duration of flowing of current to each pixel, and the equations (3) and (4) are satisfied. Since the heat capacities C.sub.th.sub.--.sub.IMG and C.sub.th.sub.--.sub.TB are proportional to the volumes of the respective cells, a desired heat capacitance C.sub.th ratio can be achieved by adjusting the area ratio, as long as the thicknesses are uniform. Even in a case where the structure is designed to satisfy the equation (5), there might be deviations from the equation (5) due to low precision in the manufacturing process or the like. In that case, the same effects as above can be achieved, as long as one of the left-hand value and the right-hand value of the equation (5) is 0.9 to 1.1 times larger than the other one of those values.

[0043] Next, an infrared signal read method is described.

[0044] In this embodiment, one reference pixel is connected to one signal line of a column of infrared detection pixels, and the reference pixels exist in a different row from those of the infrared detection pixels, as shown in FIG. 1. Since the bias voltage is applied to the row of the reference pixels at a different time from the time when the bias voltage is applied to the rows of the infrared detection pixels, signals cannot be compared at once. Therefore, differential clamp circuits that hold signals output from the reference pixels and amplify only differential signals when an infrared detection pixel is selected are used. Each of those differential clamp circuits includes a coupling capacitor 22.sub.i (i=1,2), a differential amplifier 20.sub.i (i=1, 2), a feedback switch 23.sub.i (i=1, 2), and a feedback capacitor 24.sub.i (i=1, 2).

[0045] First, during a first period, the bias voltage Vd is applied to the row select line 45.sub.0 to which the reference pixels 11.sub.1 and 11.sub.2 are connected. As a result of this, the potential of each of the vertical signal lines 44.sub.1 and 44.sub.2 becomes Vd-Vf0. This voltage value is 1.0 V, for example.

[0046] If the feedback switches 23.sub.1 and 23.sub.2 are switched on (short-circuited) at this point, a voltage follower connection is formed, and the input voltages V11, V12 and the output voltages Vo11, Vo12 of the differential amplifiers 20.sub.1 and 20.sub.2 become equal to a DC voltage V2. Here, the DC voltage V2 is a constant voltage applied to the differential amplifiers of all the columns, and may be 1.5 V, for example. At this point, the voltage of each of the coupling capacitors 22.sub.1 and 22.sub.2 on the side of the vertical signal lines 44.sub.1 and 44.sub.2 is 1.0 V, and the voltage of each of the coupling capacitors 22.sub.1 and 22.sub.2 on the side of the differential amplifiers 20.sub.1 and 20.sub.2 is 1.5 V.

[0047] When the feedback switches 23.sub.1 and 23.sub.2 are switched off (opened), the above described potential relationship is maintained, but the input voltages V11, V12 and the output voltages Vo11, Vo12 of the differential amplifiers 20.sub.1 and 20.sub.2 are cut off from each other. A current is flowed to the reference pixels 11.sub.1 and 11.sub.2 in the period from the switching on of the feedback switches 23.sub.1 and 23.sub.2 until the switching off of the feedback switches 23.sub.1 and 23.sub.2. Here, it is essential that the duration of flowing of current becomes equal to the time tsel in the equation (5).

[0048] During a second period, the bias voltage Vd is applied to the row select line 45.sub.1 to which the infrared detection pixels 12.sub.11 and 12.sub.12 are connected. As a result of this, the potential of each of the vertical signal lines 44.sub.1 and 44.sub.2 becomes Vd-(Vf0-Vsig). This voltage value is 1.001 V, for example. The difference between the potentials at both ends of each of the coupling capacitors 22.sub.1 and 22.sub.2 is kept. As the potential on the side of the vertical signal lines 44.sub.1 and 44.sub.2 is higher than that in the first period by 0.001 V, the potential of V11 (V12) increases to 1.501 V. If the gains of the differential amplifiers 20.sub.1 and 20.sub.2 are high enough at this point, the circuits formed with the coupling capacitors 22.sub.1 and 22.sub.2, the differential amplifiers 20.sub.1 and 20.sub.2, and the feedback capacitors 24.sub.1 and 24.sub.2 become integration circuits. Where Cc represents the capacity value of each of the coupling capacitors 22.sub.1 and 22.sub.2, and Cfb represents the capacity value of each of the feedback capacitors 24.sub.1 and 24.sub.2, the following relationship is established: Vo11=Cc/Cfb.times.Vsig. Where Cc is 5 pF, and Cfg=0.5 pF, for example, the relationship is expressed as Vo11=10.times.0.001=0.01 (V). As a result, the infrared signal Vsig is amplified tenfold.

[0049] As described above, this embodiment can provide an uncooled infrared imaging device that is capable of reducing the influence of the difference in self-heating temperature between the infrared detection pixels and the reference pixels.

[0050] Although series-connected diodes are used as the thermoelectric conversion elements of the infrared detection pixels and the reference pixels in the above described embodiment, the same effects as above can be achieved by using series-connected resistors.

[0051] While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel methods and systems described herein can be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein can be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.

Claims

1. An uncooled infrared imaging device comprising: a semiconductor substrate; an imaging region formed on the semiconductor substrate, a plurality of pixels being arranged in a matrix form in the imaging region, the plurality of pixels including a plurality of reference pixels arranged in at least one row, and a plurality of infrared detection pixels arranged in the remaining rows, each of the infrared detection pixels being configured to detect an incident infrared ray, each of the reference pixels including a first cell located above a first concave portion formed in a surface portion of the semiconductor substrate, the first cell including a first thermoelectric converting unit including a first infrared absorption film and a first thermoelectric conversion element, the first infrared absorption film being configured to absorb the incident infrared ray and convert the incident infrared ray into heat, the first thermoelectric conversion element being configured to convert the heat obtained by the first infrared absorption film into an electrical signal, each of the infrared detection pixels including a second cell located above a second concave portion formed in a surface portion of the semiconductor substrate, the second cell including a second thermoelectric converting unit located above the second concave portion and first and second supporting structure units supporting the second thermoelectric converting unit above the second concave portion, the second thermoelectric converting unit including a second infrared absorption film and a second thermoelectric conversion element, the second infrared absorption film being configured to absorb the incident infrared ray and convert the incident infrared ray into heat, the second thermoelectric conversion element being configured to convert the heat obtained by the second infrared absorption film into an electrical signal, the first cell having a smaller area than the second cell when viewed from an incident direction of the infrared ray; a plurality of row select lines formed in the imaging region, the plurality of row select lines corresponding to the rows of the plurality of pixels, each of the row select lines being connected to one end of the thermoelectric conversion element of each pixel of the corresponding one of the rows, each of the row select lines selecting the pixels of the corresponding one of the rows; and a plurality of signal lines formed in the imaging region, the plurality of signal lines corresponding to columns of the plurality of pixels, each of the signal lines being connected to the other end of the thermoelectric conversion element of each pixel of the corresponding one of the columns, each of the signal lines being used for reading the electrical signal from each pixel of the corresponding one of the columns.

2. The device according to claim 1, wherein, when G.sub.th.sub.--.sub.IMG represents a heat conductance of each of the infrared detection pixels, C.sub.th.sub.--.sub.IMG represents a heat capacity of each of the second cells, G.sub.th.sub.--.sub.TB represents a heat conductance of each of the reference pixels, C.sub.th.sub.--.sub.TB represents a heat capacity of each of the first cells, and tsel represents a duration of flowing of current to the infrared detection pixel and the reference pixels, the following equation is established, 1 G th_IMG { 1 - exp ( - tsel C th_IMG / G th_IMG ) } = 1 G th_TB { 1 - exp ( - tsel C th_TB / G th_TB ) } ##EQU00005## where one of the left-hand value and the right-hand value of the equation is 0.9 to 1.1 times larger than the other one of the left-hand value and the right-hand value.

3. The device according to claim 1, further comprising a plurality of differential clamp circuits corresponding to the respective signal lines, each of the differential clamp circuits including: a coupling capacitor having one terminal connected to the corresponding one of the signal lines; a differential amplifier having a positive-side input terminal connected to the other terminal of the coupling capacitor; a feedback capacitor located between the positive-side input terminal of the differential amplifier and an output terminal of the differential amplifier; and a feedback switch connected in parallel to the feedback capacitor.

4. The device according to claim 3, further comprising a plurality of read transistors corresponding to the respective signal lines, each of the read transistors having one of a source and a drain connected to the output terminal of the differential amplifier, each of the read transistors having the other one of the source and the drain connected to a read line, a gate of each of the read transistors receiving a select signal from a column select circuit.

5. The device according to claim 1, further comprising a differential amplifier configured to amplify a difference between signal voltages obtained from the infrared detection pixels and the reference pixels.

6. The device according to claim 1, wherein the first and second thermoelectric conversion elements are series-connected diodes.

7. The device according to claim 1, wherein the first and second thermoelectric conversion elements are series-connected resistors.