JP2001041818A - Bolometer type infrared ray detection element and infrared ray image sensor using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce thermal conductance for improved detecting sensitivity by allowing the thickness of a beam supporting a substrate on which, thermally separated from a base board, a thermoelectric conversion element is formed to be thicker than the substrate while the length of the beam is made equal to or longer than the element. SOLUTION: Related to an infrared ray detector, a substrate 14 is supported on a silicon base plate 12 by a pair of beams 13, and a thermoelectric conversion element 15 is formed on the substrate which is thermally separated from the base board 12. A thickness H1 of the beam 13 is thicker than the substrate 14. The rigidity of the beam 13 in the direction of base board 12 is improved by B1<=H1 where a width of the beam 13 is B1, while a length L1 of the beam 13 is made equal to or longer than an element pitch for reduced thermal conductance. On the element 15, a metal film 17 whose sheet resistance is set is laminated for preventing reflection of incident infrared ray 16. The infrared ray 16 incident on the substrate 24 is absorbed into an infrared ray absorbing layer, and a temperature change is detected through the resistance change of the element 15.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a structure of a bolometer type infrared detecting element, in particular, a beam (lead) structure of an infrared detecting diaphragm (substrate), and a two-dimensional arrangement of the elements in an infrared image sensor using the element. Output signal processing.

[0002]

2. Description of the Related Art In general, CCD cameras and photomultiplier tubes are widely used as image pickup devices, but these image pickup devices affect disturbances such as smoke, fog, sunlight and the like day and night. A far-infrared imaging camera capable of satisfying such a demand is demanded in, for example, the security market. That is, since the far-infrared camera uses far-infrared light (8 to 12 μm), the black-body radiation intensity level near normal temperature is high, and C using near- and mid-infrared light.
This is because they are superior to CD cameras and quantum infrared cameras.

There are two types of far-infrared cameras, one utilizing the quantum effect and the other utilizing the thermal effect.
Although the one utilizing the quantum effect has a high detection sensitivity, it requires a cooling (about 77 K °) mechanism, has a problem that the burden on costs and maintenance is large, and uses the uncooled thermal effect. Far infrared cameras are attracting attention.

On the other hand, such thermal effect sensors are classified into a bolometer sensor for detecting heat by resistance change, a thermopile sensor for detecting heat by a thermocouple, and a pyroelectric sensor for detecting heat by electric charge. However, in recent years, bolometer sensors utilizing the advantage of being able to be formed monolithically on a silicon wafer have been developed and manufactured.

FIG. 18 is a perspective view showing the structure of a far-infrared ray detector 1 which is a typical conventional bolometer sensor, FIG. 19 is a front view of the far-infrared ray detector 1, and FIG. FIG. 3 is a cross-sectional view taken along line AA of FIG.
This far-infrared detector 1 is disclosed in, for example, Japanese Patent No. 2710222.
No. 8 and the like, a substrate 4 is supported on a silicon substrate 2 by a pair of beams (leads) 3, and the thermoelectric conversion is performed on the substrate 4 thus thermally separated from the silicon substrate 2. An element 5 is formed. A metal film 7 having a predetermined sheet resistance is laminated on the thermoelectric conversion element 5 to prevent reflection of incident infrared rays 6. The infrared rays 6 incident on the substrate 4 are absorbed by an infrared absorbing layer (not shown) of the substrate 4 and converted into heat. The thermoelectric conversion element (bolometer material) 5 is a material whose resistance value changes with temperature.
A change in temperature that is increased by absorbing the infrared rays 6 incident on the substrate 4 is detected as a change in electric resistance. The electric resistance change can be taken out between the terminals 9 via the wiring pattern 8 formed on the beam 3 and is electrically connected to a wiring pattern on the silicon substrate 2 (not shown).

[0006]

The sensitivity Res of the far-infrared detector 1 configured as described above is expressed by the following equation. Res [V / W] = η · α · VB · (1−exp (−τi / τT)) / G (1) where η is an infrared absorption coefficient, α is a temperature coefficient of resistance, and VB Is a bias voltage, τi is an integration time, τT is a thermal time constant (= HC / G, HC is heat capacity, and G is thermal conductance).

From equation (1), in order to improve the sensitivity of the far-infrared detector 1, it is necessary to improve the infrared absorptivity η, use a bolometer material having a large resistance temperature coefficient α, and reduce the thermal conductance G. Can be considered. Further, by reducing the heat capacity HC, the thermal time constant τT can be reduced, and the responsiveness is improved.

However, since the infrared absorption coefficient η and the temperature coefficient of resistance α are determined by the material used, in order to improve the sensitivity of the far infrared detector 1,
It is necessary to reduce the thermal conductance G. here,
The thermal conductance G is expressed as follows: G = ρ · B · H / L (2) Here, ρ is the thermal conductivity, B is the width of the beam 3, H is the thickness of the beam 3, and L is the length of the beam 3.

Therefore, by using a material having a small thermal conductivity ρ, reducing the area (product of BH) of the beam 3, or increasing the length L of the beam 3, the thermal conductance G is reduced.
Can be reduced. Note that the thermal conductivity ρ is
This is the thermal conductivity of the combined structure of the beam 3 and the wiring material.

[0010] However, the thermal conductivity ρ is determined by the material used. In addition, the area of the beam 3 (B
Regarding the product of H, increasing the width of the beam 3 causes a decrease in sensor sensitivity due to a decrease in the sensor area. On the other hand, conventionally, the method of making the thickness H of the beam 3 the same as the thickness of the substrate 4 has been used. Are easily formed in the same thickness (for example, 1700 ° in Japanese Patent Application Laid-Open No. 2-196929, Nikkei Electronics 1996.5.6).
(No. 661) pp. 21 is 0.9 μm), and when the thickness H of the beam 3 is reduced, the beam 3 or the substrate 4
In this case, the influence of the distortion and warpage of the substrate 4 increases, and the thermal conductance increases due to the contact of the substrate 4 with the silicon substrate 2, thereby causing a problem of causing a device failure. Further, increasing the length L of the beam 3 may cause disturbance in the matrix arrangement of the substrate 4 formed in a rectangular shape,
There is a problem that the area (product of BH) of the beam 3 is increased.

It is an object of the present invention to provide a bolometer type infrared detecting element capable of reducing thermal conductance and improving detection sensitivity, and an infrared image sensor using the same.

[0012]

A bolometer type infrared detecting element according to the present invention is supported by a beam with respect to a substrate, and a bolometer having a thermoelectric conversion element formed on a substrate thermally separated from the substrate. Type infrared detecting element,
The thickness of the beam is made larger than the thickness of the substrate to improve the rigidity of the beam, and the length of the beam is made longer than the element length to reduce the thermal conductance.

According to the above construction, the rigidity of the beam in the film thickness direction is increased by making the thickness of the beam larger than the thickness of the substrate. Approximately the same beam stiffness in the film thickness direction is ensured, and there is no contact of the substrate with the substrate against impact or the like. And, as a result of the beam length being longer than the element length (pixel pitch), the thermal conductance can be reduced and the detection sensitivity can be improved as compared with the related art.

Further, the bolometer-type infrared detecting element according to the present invention is formed by supporting a substrate with a beam and forming a thermoelectric conversion element on a substrate thermally separated from the substrate. In the above, when the width of the beam is B and the thickness of the beam is H, the beam rigidity is improved by setting B ≦ H, and the thermal conductance is reduced by setting the beam length to be equal to or longer than the element length.

According to the above construction, the rigidity of the beam in the film thickness direction is increased by satisfying the condition of B ≦ H. Even if the beam length is longer than the element length, the beam rigidity in the film thickness direction is almost the same as the conventional one. Therefore, contact with the substrate of the substrate does not occur against impact or the like. As a result, the beam length can be made longer than the element length (pixel pitch). As a result, the thermal conductance can be reduced and the detection sensitivity can be improved.

Furthermore, in the infrared image sensor according to the present invention, the bolometer type infrared detecting element according to claim 1 or 2 is two-dimensionally arranged such that a coordinate system connecting the centers of the elements is orthogonal to each other. It is characterized by comprising.

According to the above configuration, in the infrared image sensor in which the bolometer type infrared detecting elements according to claim 1 or 2 are two-dimensionally arranged, the coordinate systems connecting the centers of the pixels of the two-dimensional pixel array are orthogonal to each other. To be placed.

Accordingly, an image can be output without distortion.

In the infrared image sensor according to the present invention, the longitudinal direction of the beam in an arbitrary element is the y-axis, the vertical direction is the x-axis, the element pitch in the y-axis direction is Py, and the element pitch in the x-axis direction is Px. Where Hoffx is the sum of the beam width and the gap interval on one side of the beam, and r1 = Py / co
sθ, r2 = Px / cos θ, θ = tan ⁻¹ (Hoff
x / Py) or θ = tan ⁻¹ (−Hoffx / Py)
In other words, the center positions of the adjacent elements on both sides of the arbitrary element in the y-axis direction are respectively converted into polar coordinates (r1, θ−π / 2) and (r1, θ + π / 2) from the center of the arbitrary element. And the center position of the adjacent element on both sides of the arbitrary element in the x-axis direction is defined by the polar coordinates (r
By arranging the elements at (2, θ) and (r2, θ + π), the respective elements are realized in a two-dimensional arrangement in which a coordinate system connecting the centers thereof is orthogonal to each other.

According to the above configuration, when the third aspect is embodied, when the bolometer type infrared detecting elements according to the first or second aspect are two-dimensionally arranged, the longitudinal direction of the beam in any element is set to the y-axis. With the vertical direction as the x-axis, the element pitch in the y-axis direction as Py, and the element pitch in the x-axis direction as Px, the center position of adjacent elements on both sides in the y-axis direction of the arbitrary element is defined as the center of the arbitrary element. 3. By arranging the beams at the polar coordinates (r1, θ−π / 2) and (r1, θ + π / 2), respectively, even if the length of the beam is equal to or longer than the element pitch Py as described in claim 1 or 2, Can be arranged without interference between them. In addition, by arranging the center positions of the adjacent elements on both sides in the x-axis direction of the arbitrary element in the polar coordinates (r2, θ) and (r2, θ + π) from the center of the arbitrary element, the center of each pixel is set. They can be arranged two-dimensionally orthogonally.

Further, in the infrared image sensor according to the present invention, the longitudinal direction of the beam in an arbitrary element is the y-axis, the vertical direction thereof is the x-axis, the element pitch in the y-axis direction is Py, and the element pitch in the x-axis direction is Py. When Px is set, the sum of the beam width and the gap interval on one side of the beam is set as Hoffx, and r1 = Py
/ Cos θ, θ = tan ⁻¹ (Hoffx / Py) or θ = tan ⁻¹ (−Hoffx / Py), the center position of the adjacent element on both sides in the y-axis direction of the arbitrary element is determined by the arbitrary Polar coordinates (r1, θ-π from the center of the element
/ 2), (r1, θ + π / 2), and the center positions of adjacent elements on both sides of the arbitrary element in the x-axis direction are respectively set to polar coordinates (Px, 0), (P
x, π), multiplying the output of the element adjacent in the y-axis direction by a predetermined weight, and calculating the complemented output, thereby connecting each of the elements in a pseudo-coordinate system. Realize a two-dimensional arrangement orthogonal to each other.

According to the above configuration, in implementing the third aspect, when the bolometer type infrared detecting elements according to the first or second aspect are two-dimensionally arranged, they are located in the longitudinal direction of the beam in an arbitrary element. 2. The method according to claim 1, wherein the center positions of adjacent pixels on both sides are offset in directions opposite to each other in a direction perpendicular to the longitudinal axis of the beam by a sum Hoffx of a beam (lead) width and a gap interval on both sides of the beam. Or 2
Such a beam having an element pitch or more is arranged without interference, whereby the axis connecting the centers of adjacent pixels in a direction intersecting with the longitudinal axis of the beam is orthogonal to the longitudinal axis of the beam. And an offset is generated from the rectangular coordinates.
For example, when arranging two-dimensional pixels, an image in which the center position of an adjacent element on both sides of a certain element is offset in a direction orthogonal to each other in the opposite direction by the beam width and the gap interval on both sides of the beam is displayed (display). When the horizontal signal axis is set, the position of each element is shifted in the horizontal direction between adjacent horizontal scanning lines, so that the output of the adjacent element is multiplied by a predetermined weight and added. A signal equivalent to the signal at the original position, that is, the signal at the original position in the rectangular coordinate system is calculated, and the deviation is corrected by the present signal.

In the infrared image sensor according to the present invention, the output of an arbitrary element is expressed by S (m, n) (m is the x coordinate, n
Is the y-coordinate), the complementary output F multiplied by the weight
(M, n) is expressed as F (m, n) = ｛S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n-1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ or F (m, n) = {S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n + 1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ where MOD (a, b) is a remainder obtained by dividing a by b, Q
UOTIENT (a, b) is characterized in that it is calculated from a quotient obtained by dividing a by b.

According to the above-described structure, the fifth embodiment is embodied, and a complementary output F (m) for display on a display in a rectangular coordinate system is provided by using a two-dimensional element output which is not actually in the rectangular coordinate system. , N) can be calculated.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described.
The following is a description based on FIGS. 1 to 5 and FIGS. 19 and 20.

FIG. 1 is a sectional view of a far-infrared detector 11 which is a bolometer sensor according to an embodiment of the present invention.
FIG. 2 is a front view of the far-infrared detector 11. 2, the section line in FIG. 1 is indicated by BB. The far-infrared detector 11 includes a substrate (diaphragm) 1 on a silicon substrate 12 by a pair of beams (leads) 13.
4 is supported on the substrate 14 thus thermally separated from the silicon substrate 12.
Are formed.

On the thermoelectric conversion element 15, incident infrared rays 16
A metal film 17 having a predetermined sheet resistance is laminated to prevent reflection of the light. The infrared light 16 incident on the substrate 14 is absorbed by an infrared absorbing layer (not shown) of the substrate 14 and converted into heat. The thermoelectric conversion element 15 is made of, for example, an oxide of a transition metal, Pt or Tt.
It is made of a bolometer material such as i, and has a resistance value that changes with temperature. The substrate 14 detects an increased temperature change by absorbing the incident infrared light 16 as an electric resistance change.

The change in electric resistance is taken out via a wiring pattern 18 formed on the beam 13 and given to a wiring pattern on a silicon substrate 12 (not shown). On the silicon substrate 12, facing the substrate 14,
A reflection film 19 for reflecting infrared light transmitted through the substrate 14 is formed.

For example, the substrate 14 and the beam 13 are made of SiO ₂ or SiN, and the wiring pattern 18 and the reflection film 19 are made of Al. The wiring pattern 18 may be made of a material other than Al, such as Ti. The sheet resistance of the metal film 17 is:
For example, it is 377Ω / □. The height K of the substrate 14 from the silicon substrate 12 (reflection film 19) is selected to be the wavelength of the infrared light 16, for example, 1/4 of 10 μm, and the phase of the reflected wave from the reflection film 19 is applied to the reflection film 19. By inverting the phase of the output wave, reflection on the surface of the substrate 14 is suppressed.

It should be noted that, in the present invention, the thickness H1 of the beam 13 (including the wiring pattern 18 in FIG. 1) is larger than the thickness of the substrate 14. As a result, B1 ≦ H1 with respect to the width B1 of the beam 13 as described later, the rigidity of the beam 13 in the direction of the substrate 12 is improved, and the length L1 of the beam 13 is set to be equal to or more than the element pitch Py, and 1 is to reduce the thermal conductance G and improve the sensitivity Res.

FIG. 3 is a cross-sectional view for explaining the step of forming the far-infrared detector 11 for forming the beam 13 thicker than the substrate 14 as described above. In FIG. 3, the reflection film 19 is omitted. First, FIG.
As shown by, a sacrifice layer 21 made of an organic material such as polyimide or an inorganic material such as amorphous silicon is formed on the silicon substrate 12.

Next, as shown in FIG.
The substrate 14 and the lower insulating layer 22 which becomes the beam 13 are formed by O ₂ or SiN, and a portion corresponding to the substrate 14 is indicated by a reference numeral 23.
Etching is performed by a predetermined amount by dry etching or wet etching such as RIE or ion milling.

Subsequently, as shown in FIG. 3C, the thermosensitive layer 24 serving as the thermoelectric conversion element 15, Ti, Ta or A
An electrode layer 25 which is made of a metal material such as 1 and overlaps a part of the temperature sensitive layer 24 and serves as the wiring pattern 18 and an upper insulating layer 26 not shown in FIG. 1 are sequentially laminated. The temperature-sensitive layer 2 of the electrode layer 25
4 is processed into a desired shape (the processed shape is not shown).

Further, as shown in FIG.
The periphery of the substrate 14 and the beam 13 is etched down to the sacrificial layer 21 by dry etching or wet etching such as IE or ion milling. Then, as shown in FIG.
As shown by, when the sacrificial layer 21 between the substrate 14 and the beam 13 and the silicon substrate 12 is removed, the far-infrared detector 11 is completed.

In the above-described example, the lower insulating layer 22 to be the beam 13 and the substrate 14 is uniformly formed, and the portion corresponding to the substrate 14 is selectively etched, whereby the beam 13 and the substrate 14 are formed. Although the respective layers 14 are formed to have appropriate thicknesses, the film may be formed uniformly to the thickness of the substrate 14 and only the beam 13 may be selectively formed to a predetermined beam thickness by lift-off or the like. .

Here, parameters related to the present invention in the far-infrared detector 1 will be described with reference to FIGS. The width of the beam 3 is B0, the thickness is H0, and the length is L0. Therefore, the sectional area A0 of the beam 3 is B0
× H0. The longitudinal direction of the beam 3 is defined as the y-axis, the vertical direction is defined as the x-axis, the element pitch in the y-axis direction is defined as Py, and the element pitch in the x-axis direction is defined as Px. The thermal conductivity of the beam 3 [rho is reveal the sum of the thermal conductivity [rho _{the Wiring} by thermal conductivity [rho _SiO2 and the beam of the wiring pattern 8 by the beam structure.

In this conventional far-infrared detector 1, the length L0 of the beam 3 is L0 <Px as compared with the element pitch Px. Further, the beam width B0 and the beam thickness H0 have a relationship of B0> H0. In this structure, the beam 3 has a cantilever structure,
The load W of the substrate 4 may be applied to the tip. Moment of inertia I0x of deflection of the beam 3 at this time is ^{I0x = B0 · H0 3/12} ... (3), also when the E the Young's modulus of the beam 3, the maximum deflection V0max of the beam 3 is generated at the tip V0max = W · L0 / (3 · E · I0x) (4)

Next, assuming that the sectional area of the beam 3 is A0, the width is B0, the thickness is H0, the length is L0, and the thermal conductivity of the beam structural material is ρ, the thermal conductance G0 is as follows. From Equation 2, G0 = ρ · B0 · H0 / L0 = ρ · A0 / L0 (5)

On the other hand, the far infrared detector 1 of the present invention
In 1, the substrate 14 has the same material and the same shape as the substrate 4, and the element pitch is the same for Py and Px. However, as described above, the length L1 of the beam 13 is L1> Px compared to the element pitch Px, and the width B1 and the thickness H1 of the beam 13 have a relationship of B1 <H1. The cross-sectional area A1 of the beam 13 is B1 × H1.

Also in the structure of the present invention, the beam 13 may have a cantilever structure, and the load W of the substrate 14 may be applied to the tip. Moment of inertia I1x deflection of the beam 13 at this time ^{is, I1x = B1 · H1 3/} 12 a ... (6), also when the E the Young's modulus of the beam 13, the beam 13
Is generated at the tip, and V1max = W · L1 / (3 · E · I1x) (7)

Next, the thermal conductance G1 is such that the cross-sectional area of the beam 13 is A1, the width is B1, the thickness is H1, and the length is L1.
When the thermal conductivity of the beam structural material is ρ, from the above equation 2, G1 = ρ · B1 · H1 / L1 = ρ · A1 / L1 (8)

Here, the sectional areas of the beams 3 and 13 are the same,
That is, A0 = A1 is set, and the maximum allowable movements at the tips are the same, that is, V0max = V1max.
First, G0 / G1 = L0 / L1 (9) is obtained from Expressions 5 and 8. On the other hand, from Equations 4 and 7, W · L0 / (3 · E · I0x) = W · L1 / (3 · E · I1x) (10) Therefore, L0 / I0x = L1 / I1x (11) ), and further from equation 3 and equation 6 ^{which, L0 / (B0 · H0 3} /12) = L1 / (B1 · H1 3/12) ... (12) ^{becomes, L0 / (A0 · H0 2} ) = L1 / (A1 · H1 ² ) (13) From the above-mentioned A0 = A1 and the above equation 9, (G0 / G1) = (H1 / H0) ² (14)

That is, the sectional areas A0, A1 of the beams 3, 13
Are the same, the maximum values V0max and V1max allowed at the tips of the beams 3 and 13 are the same.
Irrespective of the lengths L0 and L1, the ratio between the thermal conductances G0 and G1 is (H1 / H0) ² . Therefore,
The thickness of the beam 13 is made thicker than H0 in a certain state (H0
/ H1 <1), the thermal conductance G1 is increased from G0 by (H
0 / H1) ² can be reduced. The relationship of Equation 14 is shown in the graph of FIG.

In the following, when the sectional area A1 and the length L1 of the beam 13 are constant, that is, when the thermal conductance G1 is constant, the thickness of the beam 13 is changed from H1 to H2 and the width of the beam 13 is changed from B1 to B2. The tip deflection V1 of the beam 13 when it is made to
The relationship between max and V2max will be described.

The rigidity of the beam 13, similarly to the equation 3 and equation 6, since represented by ^{I2x = B2 · H2 3/12} = A2 · H2 2/12 ... (15), the same cross-sectional area A1 = A2 In this case, the moment of inertia I2x is represented by the square of the thickness H2. The deflection V2max at the tip of the beam is obtained by substituting the equation 15 from the equation V2max = W · L1 ³ / (3 · E · I2x) (16), as in the equations 4 and 7, V2max = 4W · L1 ³ / (E · B2 · H2 ³ ) (17) Similarly, V1max = 4W · L1 ³ / (E · B1 · H1 ³ ) (18) Here, since A1 = B1 · H1 = B2 · H2 (19), V1max / V2max = (H2 / B2) (B1 / H1) (20) is obtained.

Then, if H2 = B2 is set for comparison, the above equation (20) becomes: V1max / V2max = (B1 / H1) (21) Therefore, the deflection V2 when H2 = B2
The ratio of the deflection V1max to the maximum is (B1 / H
It can be expressed as a ratio of 1). This relationship is shown in the graph of FIG. That is, when the cross-sectional area A1 and the length L1 of the beam 13 are constant, and therefore, the thermal conductance G1 is constant, the tip deflection V1max of the beam 13 is equal to the width B1 of the beam 13.
And the thickness H1. As B1 <H1,
It is understood that the deflection V1max is reduced.

As described above, according to the present invention, the thickness H1
Is larger than the thickness of the substrate 14, and the width B1 of the beam 13 is set to be B1 ≦ H1 with respect to the thickness H1.
By increasing the rigidity of the beam 13 in the thickness direction,
Even when the length L1 is equal to or longer than the element length Py, the beam rigidity in the film thickness direction is substantially the same as that of the related art, and the substrate 14 does not come into contact with the silicon substrate 12 in response to impact or the like. The thermal conductance G1 can be reduced, and the detection sensitivity Res can be improved. However, in consideration of the impact resistance in the lateral direction, it is not desirable to make B1≪H1 unnecessarily.

Another embodiment of the present invention will be described below with reference to FIGS. 6 and 7.

FIG. 6 is a sectional view of a far-infrared detector 31 which is a bolometer sensor according to another embodiment of the present invention.
This far-infrared detector 31 is the far-infrared detector 11 described above.
, And corresponding parts are denoted by the same reference numerals and description thereof is omitted. It should be noted that the far-infrared detector 31 has the same height as the lower surface of the beam 13 and the substrate 14, whereas the far-infrared detector 11 has the same height as that of the beam 33 and the substrate 14. That is, the heights of the upper surfaces are aligned.

FIG. 7 shows a far infrared detector 31 as described above.
FIG. 6 is a cross-sectional view for describing a manufacturing step. FIG. 7 corresponds to FIG. First, as shown in FIG. 7A, the sacrificial layer 21 is formed on the silicon substrate 12, and a portion corresponding to the beam 33 is dry-etched by RIE or ion milling or the like as indicated by reference numeral 34. Etching is performed by a predetermined amount by wet etching.

Next, as shown in FIG. 7B, the lower insulating layer 22 to be the substrate 14 and the beam 33 is formed. Further thereon, the temperature-sensitive layer 24 and the electrode layer 2
5 and the upper insulating layer 26 are stacked.

Subsequently, as shown in FIG. 7C, the unevenness of the upper insulating layer 26 is flattened by etching back or grinding by CMP, and as shown in FIG. The periphery of the substrate 14 and the beam 33 is etched to the sacrifice layer 21 by dry etching such as milling or wet etching. Then, by oxygen plasma or aqueous solution,
As shown in FIG. 7E, when the sacrificial layer 21 between the substrate 14 and the beam 33 and the silicon substrate 12 is removed, the far-infrared detector 31 is completed.

Regarding still another embodiment of the present invention,
The following is a description based on FIG.

FIG. 8 is a sectional view of a far-infrared ray detector 41 which is a bolometer sensor according to still another embodiment of the present invention. This far-infrared detector 41 is similar to the far-infrared detectors 11 and 31 in that the thickness H1 of the beam 13 is set to B1 ≦ H1 with respect to the width B1, but the thickness of the substrate 44 is 13 is formed.

Therefore, although the heat capacity HC of the substrate 44 is larger than that of the substrate 14, if the sensitivity and the time constant can be tolerated, this configuration can be easily performed by a conventional manufacturing process.

Another embodiment of the present invention will be described below with reference to FIG.

FIG. 9 is a front view of an infrared image sensor 51 according to another embodiment of the present invention. The infrared image sensor 51 includes the above-described far infrared detectors 11, 31,
The device is formed by arranging elements consisting of any one of 41 in a two-dimensional matrix. Each element is indicated by a reference numeral S, and is further indicated by an address in parentheses (when generically indicated, only S is indicated). In FIG. 9, 3 × 3 elements S (m−1, n−1), S are centered on element S (m, n).
(M, n−1), S (m + 1, n−1); S (m−1,
n), S (m, n), S (m + 1, n); S (m-1,
n + 1), S (m, n + 1), and S (m + 1, n + 1). The coordinate system is indicated by x and y.

For the central element S (m, n), x
The elements S (m ± 1, n) adjacent in the axial direction are arranged with their center positions offset by Hoffy in the y-axis direction in order to make the length L1 of the beam 13 longer than the element pitch Px. The offset Hoffy is equal to the beam width B1.
Hoffy = B1 + D1... (22) The gap distances D1 are equal to each other and equal to the distance between the beam 13 and the substrate 14.

Therefore, in the example of FIG. 9, nine elements S
Are not orthogonal to each other. For this reason, when two-dimensional detection is performed using the matrix configuration of FIG. 9 and output as an image, the image is distorted if the display (for example, a cathode ray tube or a liquid crystal display) has straight lines connecting the center positions of the pixels on the image output side. Although it becomes an image, it may be output as it is in an application in which there is no problem in displaying it on the rectangular coordinate system. Further, since the offset Hoffy between the elements S is known,
When performing an operation such as image extraction, a known offset Hoff is used.
An operation such as image extraction may be performed in consideration of y.

Regarding still another embodiment of the present invention,
The following is a description based on FIG.

FIG. 10 is a front view of an infrared image sensor 61 according to still another embodiment of the present invention. This infrared image sensor 61 has the same configuration as the infrared image sensor 51 described above, and it should be noted that the x and y coordinates are exchanged with each other. In such an x, y coordinate system, when the elements S whose length L1 is longer than the element pitch Px are two-dimensionally arrayed as described above, the y-axis is positioned with respect to the center element S (m, n). The center positions of the elements S (m, n ± 1) on both sides adjacent to each other in the X direction are H
Offset by offset. Hoffx corresponds to FIG.
Similarly, Hoffx = B1 + D1 (23).

Here, r1 = Py / cos θ (24) θ = tan ⁻¹ (−Hoffx / Py) (25) If the element S (m, n) is shifted from the center of the element S (m, n),
The center of -1) is located at polar coordinates (r1, θ-π / 2),
The center of the element S (m, n + 1) is at the polar coordinates (r1, θ + π /
It is located in 2). Note that θ is not shown in FIG. When the direction in which the beams 13 are pulled out is opposite, θ = tan ⁻¹ (Hoffx / Py) (26) is used instead of the equation 25.

Further, for the element S (m, n), x
The elements S (m-1, n) and S (m
+1 and n) are the polar coordinates (Px, 0) and
(Px, π).

With this configuration, in the rectangular coordinate display, each element S is on each x-axis, and the offset Hoffx occurs in the x-axis direction. That is, this corresponds to the occurrence of an offset of each element S in the horizontal scanning line direction when displaying on the orthogonal display. Since the offset Hoffx of the element in the horizontal direction between the adjacent scanning lines is known, a predetermined weight is multiplied from the detection signal of the adjacent element, and both are added to obtain a pixel position corresponding to the orthogonal coordinate system. Can be calculated.

For example, as shown in FIG.
(M, n) [m = 0, 1,..., 11, n = 0, 1,.
9] at pitches Px, Py, x in the x and y axis directions, respectively.
The offset amounts in the axial direction are arranged in a matrix with Hoffx (where S (m, n) <S (m, n + 1),
In other words, it indicates that the element S (m, n + 1) is relatively rightward on the x-axis than the element S (m, n). The element S (j, -1) [j = 1 to 4, 6 to 9] is a dummy element in weight calculation described later. ) And the element S
Complementary output F obtained by multiplying the output of (m, n) by weight
(M, n) is: F (m, n) = {S (m, n) * (Px-MOD (m, (QUOTIENT (Px, Hoffx)) * Hoffx) / Px} + {S (m, n) -1) * MOD (m, (QUOTIENT (Px, Hoffx)) * Hoffx) / Px｝ (27) where MOD (a, b) is a
Is divided by b, QUAOTIENT (a, b) is a
Divided by b.

Tables 1 and 2 show the calculation results of this example.
Table 1 shows the element output of the element S (m, n) in FIG. 11 as it is, and Table 2 shows the complementary output F (m, n) of the position of the rectangular coordinate system weighted using the above equation (27). It is a calculation result of. However, as shown in FIG.
x: Hoffx = 5: 1. Also, n =
The -1 dummy column is a dummy output used when performing weight complementation of n = 0 in Expression 27. In this example, the output “0” is used for convenience, but the intermediate value of the element output, the maximum value, Any value, such as a minimum value, may be used depending on the application.

[0067]

[Table 1]

[0068]

[Table 2]

Next, the element S (m, n) and the element S (m, n)
+1) is S (m, n)> S (m, n +
1), that is, the element S (m, n + 1) is
An example of the case where the position is relatively left on the x-axis relative to n) will be described with reference to FIG. 12, which is a diagram similar to FIG. The element S (j, 12) [j = 1 to 4, 6 to 9] is the dummy element. In this case, a complementary output F (m, n) obtained by multiplying the output of the element S (m, n) by a weight is given by F (m, n) = S (m, n) * (Px−MOD (m , (QUOTIENT (Px, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n + 1) * MOD (m, (QUOTIENT (Px, Hoffx)) * Hoffx) / Px｝ (28) be able to.

Tables 3 and 4 show the calculation results of this example.
Table 3 shows the element output of the element S (m, n) of FIG. 12 as it is, and Table 2 shows the complementary output F (m, n) of the position of the rectangular coordinate system weighted by using the equation (28). It is a calculation result of. However, similarly to FIG. 11, Px: Hoff
x = 5: 1. In the dummy column of n = 12 in Table 3, the output “0” is used for convenience, but what value, such as an intermediate value, a maximum value, and a minimum value of the element output, is adapted to the intended use. May be used.

[0071]

[Table 3]

[0072]

[Table 4]

By performing the weighting operation as described above, when displaying on the display of the rectangular coordinate system, the complementary output F at the pixel position corresponding to the rectangular coordinate system is displayed.
(M, n) can be calculated.

FIG. 13 shows another embodiment of the present invention.
15 will be described below.

FIG. 13 is a front view of an infrared image sensor 71 according to another embodiment of the present invention. The infrared image sensor 71 includes the infrared image sensor 51,
It is noteworthy that each element S is arranged such that its axes Lx, Ly passing through its center are orthogonal to each other and parallel to their axes Lx, Ly, x ',
That is, the y 'coordinate is set.

That is, the elements S (m−1, n−1), S (m−1, n), S (m−) are located on the axis Ly (m−1).
1, n + 1) are arranged, and the element S is located on the axis Ly (m).
(M, n-1), S (m, n), S (m, n + 1) are arranged, and the element S (m + 1, n) is located on the axis Ly (m + 1).
-1), S (m + 1, n) and S (m + 1, n + 1), and similarly, the element S (m-
1, n-1), S (m, n-1), S (m + 1, n-
1) is arranged, and the element S (m−) is disposed on the axis Lx (n).
1, n), S (m, n) and S (m + 1, n) are arranged, and the element S (m−1, n +) is located on the axis Lx (n + 1).
1), S (m, n + 1), and S (m + 1, n + 1).

Referring to FIG. 14, the x 'and y' coordinates are rotated by θ from the x and y coordinates. Each element S
In order to set the center of (m, n) to C (m, n) and use the above-described rectangular coordinate system, for example, the element S (m + 1, n) is replaced by the element S (m, n). m, n) center C
The axis Lx (n) inclined from the x-axis by the inclination θ of the axis Ly (m) with respect to the axis y (m, n) passing through the (m, n) and parallel to the y-axis, and the center C (m + 1) , N), the center C ′ (m + 1, n) may be arranged at the intersection with the axis y (m + 1, n) parallel to the y-axis. That is, the center C ′ (m + 1, n) of the element S (m + 1, n)
n) is a center C (m, m) on the axis y (m + 1, n).
It may be arranged at a position rotated by θ from the center C (m + 1, n) around n).

The center C 'of the element S (m + 1, n)
The reason why (m + 1, n) is located on the axis y (m + 1, n) is that the element pitch Px is required as a gap interval between the adjacent element S (m, n). Here, the moving distance Hoffy (C (m + 1, n) and C ′ (m + 1,
n)) is represented by Hoffy = Px × tan θ (29). Here, θ = tan ⁻¹ (Hoffx / Py) (26) (Hoffx and Py are described in FIG. 10).

FIG. 15 shows an element S on the axis Ly (m).
Based on (m, n-1), S (m, n), and S (m, n + 1), adjacent elements S (m + 1, n-)
1), S (m + 1, n) and S (m + 1, n + 1) are displaced as described above and arranged on the axis Ly (m + 1), and the center of each element is orthogonal coordinates of x ′, y ′ coordinates. This shows a state in which it is positioned in the system.

Here, the element pitches Px ′ and Py ′ in the new orthogonal coordinate system of the x ′ and y ′ coordinates are respectively Px ′ = Px / cos θ (30) Py ′ = Py / cos θ (31) Is represented as

In this case, the center C (m, n-1) of the element S (m, n-1) adjacent to the element S (m, n) in the y-axis direction
Is the polar coordinate (r1, θ + π / 2) from the center C (m, n).
Located in. The center C (m, n + 1) of another element S (m, n + 1) adjacent in the y-axis direction is the center C
It is located at polar coordinates (r1, θ-π / 2) from (m, n). Here, r1 = Py / cos θ (24) θ = tan ⁻¹ (−Hoffx / Py) (25)

The center C ′ (m + 1, n) of the element S (m + 1, n) adjacent to the element S (m, n) in the x-axis direction
Is located at polar coordinates (r2, θ) from the center C (m, n). Further, another element S (m−
1, n) (see FIG. 13) C ′ (m−1, n)
Is located at polar coordinates (r2, θ + π) from the center C (m, n). Here, r2 = Px / cos θ
... (32) θ = tan ^-1 (−
Hoffx / Py) ... (2
5).

By arranging the adjacent elements in this manner, a matrix arrangement in the rectangular coordinate system shown in FIG. 13 can be realized, and a complicated arrangement like the infrared image sensor 61 shown in FIGS. The display on the display in the rectangular coordinate system can be performed without performing the signal processing.

[0084] In still another embodiment of the present invention,
The following is a description based on FIG.

FIG. 16 is a front view of an infrared image sensor 81 according to still another embodiment of the present invention. This infrared image sensor 81 is similar to the above-described infrared image sensor 51. It should be noted that although the length L1 of the beam 13 in the infrared image sensor 51 is about 1.5 times the element pitch Px. On the other hand, the length L2 is the element pitch P
x is formed about 2.5 times.

The center C (m ± 1, n) of the element S (m ± 1, n) adjacent in the x-axis direction to the center C (m, n) of the element S (m, n) is as described above. Infrared image sensor 51
And is offset by Hoffy in the y-axis direction. However, the element pitch in the y-axis direction has expanded from Py to Py ″.
4, the width of B10, the gap between adjacent elements and the beam 1
Assuming that the distance between the substrate 3 and the substrate 14 is D1, Py "= B10 + 4.times.B1 + 5.times.D1 (34) with respect to Py = B10 + 3.times.B1 + 4.times.D1 (34), and Py" -Py = B1 + D1 (34) 33) It is only necessary to increase the element pitch.

Thus, the length L2 of the beam 13 can be extended, and the thermal conductance G can be further reduced. Here, the element pitch Px in the x-axis direction is the same as that of the infrared image sensor 51 except for the element pitch Py ″ in the y-axis direction, and the infrared image sensor 51 shown in FIG. The concept of converting x, y coordinates into x ′, y ′ coordinates with respect to the infrared image sensor 61 or the infrared image sensor 71 shown in FIGS. 13 to 15 can be applied to the infrared image sensor 81.

The far-infrared detectors 11, 31,
In each of the infrared image sensors 51, 61, 71, and 81 having a two-dimensional matrix array of 41, a signal detected by each element is represented by a detection value at a center position of each element when represented by a two-dimensional coordinate system. Become. Therefore, as shown by hatching in FIG. 17, the area 13a outside the element pitches Px and Py of the beam 13 may or may not be used as the infrared detection range.

When not used, for example, Japanese Unexamined Patent Application Publication No.
As described in Japanese Patent Application Laid-Open No. 96929, the sheet resistance of the surface metal film and the fact that a thick conductive layer is formed in parallel at a position λ / 4 away from the wavelength λ to take advantage of the fact that the infrared absorptivity is different, The infrared absorptance of the area 13a may be reduced.

[0090]

As described above, the bolometer-type infrared detecting element according to the present invention is formed by forming a thermoelectric conversion element on a substrate that is supported by a beam on a substrate and thermally separated from the substrate. In the bolometer type infrared detecting element,
The thickness of the beam is made larger than the thickness of the substrate to improve the rigidity of the beam, and the length of the beam is made longer than the element length to reduce the thermal conductance.

Therefore, the detection sensitivity can be improved.

Further, the bolometer type infrared detecting element according to the present invention is supported by the beam with respect to the substrate as described above.
In a bolometer-type infrared detecting element in which a thermoelectric conversion element is formed on a substrate thermally separated from the substrate, when the width of the beam is B and the thickness of the beam is H, B ≦ H
The beam rigidity is improved, and the thermal conductance is reduced by setting the length of the beam to be equal to or longer than the element length.

Therefore, the detection sensitivity can be improved.

Further, as described above, the infrared image sensor according to the present invention is arranged such that the bolometer type infrared detecting element according to claim 1 or 2 is arranged such that the coordinate systems connecting the centers of the elements are orthogonal to each other. It consists of a two-dimensional arrangement.

Therefore, an image can be output without distortion.

Further, in implementing the infrared image sensor according to the present invention as described above,
When the bolometer type infrared detecting elements according to claim 1 or 2 are two-dimensionally arranged, the longitudinal direction of a beam in an arbitrary element is the y-axis, the vertical direction is the x-axis, the element pitch in the y-axis direction is Py, and the x-axis. Assuming that the element pitch in the direction is Px, the center position of the adjacent element on both sides of the arbitrary element in the y-axis direction is defined by the polar coordinates (r1, θ−π /
2), (r1, θ + π / 2), and the center position of the adjacent element on both sides in the x-axis direction in the arbitrary element is
The polar coordinates (r2, θ) from the center of the arbitrary element,
(R2, θ + π).

Therefore, in the y-axis direction, even if the length of the beam is equal to or longer than the element pitch Py as in the first or second aspect, the beams can be arranged without interference between adjacent elements. In addition, the x
In the axial direction, the centers of the pixels can be two-dimensionally arranged orthogonally.

Further, as described above, the infrared image sensor according to the present invention, when embodying the third aspect, has a structure in which the bolometer type infrared detecting elements according to the first or second aspect are two-dimensionally arranged. The center position of the adjacent pixels on both sides located in the longitudinal direction of the beam in the device of the above is determined by the sum Hof of the beam (lead) width and the gap interval on both sides of the beam.
fx, and are arranged offset from each other in the direction perpendicular to the longitudinal axis of the beam in directions opposite to each other, and adjacent pixels on both sides of the arbitrary element located in the direction perpendicular to the longitudinal axis of the beam remain as they are. By multiplying the output of the element by a predetermined weight and calculating the complemented output, the two-dimensional arrangement of the respective elements is realized in a pseudo manner in a coordinate system connecting the centers thereof.

Therefore, it is possible to arrange beams having a pitch equal to or larger than the element pitch without interference, and it is possible to correct a deviation from the orthogonal coordinates.

In the infrared image sensor according to the present invention, as described above, in realizing the fifth aspect, the output of an arbitrary element is represented by S (m, n) (m is the x coordinate, n
Is the y-coordinate), the complementary output F multiplied by the weight
(M, n) is expressed as F (m, n) = ｛S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n-1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ or F (m, n) = {S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n + 1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ where MOD (a, b) is a remainder obtained by dividing a by b, Q
UOTIENT (a, b) is calculated from the quotient obtained by dividing a by b.

Therefore, a complementary output F (m, n) for display on a display in a rectangular coordinate system can be calculated using a two-dimensional element output not in the rectangular coordinate system.

[Brief description of the drawings]

FIG. 1 is a cross-sectional view of a far-infrared detector that is a bolometer sensor according to an embodiment of the present invention.

FIG. 2 is a front view of the far-infrared detector shown in FIG.

FIG. 3 is a cross-sectional view for explaining a manufacturing process of the far-infrared detector shown in FIGS. 1 and 2.

FIG. 4 is a graph for explaining a change in thermal conductance with respect to a change in beam thickness in the bolometer sensor.

FIG. 5 is a graph for explaining a change in deflection of the beam with respect to a change in the ratio of the width to the thickness of the beam in the bolometer sensor.

FIG. 6 is a cross-sectional view of a far-infrared detector which is a bolometer sensor according to another embodiment of the present invention.

FIG. 7 is a cross-sectional view for explaining a step of producing the far-infrared detector shown in FIG.

FIG. 8 is a cross-sectional view of a far-infrared detector which is a bolometer sensor according to still another embodiment of the present invention.

FIG. 9 is a front view of an infrared image sensor according to another embodiment of the present invention.

FIG. 10 is a front view of an infrared image sensor according to still another embodiment of the present invention.

11 is a schematic front view for explaining a method of converting element outputs into data of a rectangular coordinate system in the infrared image sensor shown in FIG.

12 is a schematic front view for explaining a method of converting element outputs into data of a rectangular coordinate system in the infrared image sensor shown in FIG.

FIG. 13 is a front view of an infrared image sensor according to another embodiment of the present invention.

FIG. 14 is a front view for explaining a method of arranging elements in the infrared image sensor shown in FIG.

FIG. 15 is a front view for explaining a method of arranging elements in the infrared image sensor shown in FIG.

FIG. 16 is a front view of an infrared image sensor according to still another embodiment of the present invention.

FIG. 17 is a front view for explaining an infrared detection range of each of the far infrared detectors.

FIG. 18 is a perspective view showing a structure of a far-infrared detector which is a typical conventional bolometer sensor.

FIG. 19 is a front view of the far-infrared detector shown in FIG.

FIG. 20 is a sectional view taken along the line AA of FIG. 19;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Far infrared detector 12 Silicon substrate 13 Beam (lead) 14 Substrate (diaphragm) 15 Thermoelectric conversion element 16 Infrared 17 Metal film 18 Wiring pattern 19 Reflection film 21 Sacrificial layer 22 Lower insulating layer 24 Temperature sensing layer 25 Electrode layer 26 Upper part Insulating layer 31 Far-infrared detector 33 Beam (lead) 41 Far-infrared detector 44 Substrate (diaphragm) 51 Infrared image sensor 61 Infrared image sensor 71 Infrared image sensor 81 Infrared image sensor B1 Beam width B10 Width of substrate C (m , N) Center C '(m, n) Center D1 Gap interval between adjacent elements H1 Beam thickness Hoffx offset Hoffy offset L1 Beam length L2 Beam length S (m, n) Element Px Element pitch Px' Element pitch Py Element Pitch Py 'element Pitch Py ”Element pitch

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G065 AA04 AB02 BA12 BA34 CA13 CA27 4M118 AA01 AA05 AA10 CA14 CA35 CB14 EA01 GA10 GD15

Claims (6)

[Claims] 1. A bolometer-type infrared detecting element comprising a thermoelectric conversion element formed on a substrate supported on a substrate and thermally separated from the substrate by a beam, wherein the thickness of the beam is controlled by the substrate. A bolometer-type infrared detection element characterized in that the thickness is larger than the thickness of the element to improve the rigidity of the beam, and the length of the beam is longer than the element length to reduce the thermal conductance. 2. A bolometer-type infrared detecting element comprising a thermoelectric conversion element formed on a substrate thermally separated from the substrate and supported by a beam with respect to the substrate. A bolometer-type infrared detecting element characterized in that when the thickness is H, the beam rigidity is improved by setting B ≦ H, and the thermal conductance is reduced by setting the beam length to the element length or more. 3. An infrared image sensor, wherein the bolometer-type infrared detecting elements according to claim 1 or 2 are two-dimensionally arranged such that a coordinate system connecting the centers of the elements is orthogonal to each other. 4. The longitudinal direction of a beam in an arbitrary element is defined as a y-axis,
The vertical direction is the x-axis, and the element pitch in the y-axis direction is Py, x
When the element pitch in the axial direction is Px, the sum of the beam width and the gap interval on one side of the beam is Hoffx, and r1 = Py / cos θ r2 = Px / cos θ θ = tan ⁻¹ (Hoffx / Py) or θ = tan ⁻¹ (−Hoffx / Py), the center position of the adjacent element on both sides of the arbitrary element in the y-axis direction can be calculated from the center of the arbitrary element in polar coordinates (r1,
θ-π / 2) and (r1, θ + π / 2), and the center position of the adjacent element on both sides in the x-axis direction of the arbitrary element is set to polar coordinates (r2,
4. The infrared image sensor according to claim 3, wherein by arranging the elements at (θ) and (r2, θ + π), the respective elements realize a two-dimensional arrangement in which a coordinate system connecting the centers thereof is orthogonal to each other. 5. The y-axis means the longitudinal direction of a beam in an arbitrary element;
The vertical direction is the x-axis, and the element pitch in the y-axis direction is Py, x
When the element pitch in the axial direction is Px, the sum of the beam width and the gap interval on one side of the beam is Hoffx, and r1 = Py / cos θ θ = tan ⁻¹ (Hoffx / Py) or θ = tan ⁻¹ (− Hoffx / Py), the center positions of adjacent elements on both sides of the arbitrary element in the y-axis direction are respectively set to polar coordinates (r1,
θ-π / 2) and (r1, θ + π / 2), and the center positions of the adjacent elements on both sides in the x-axis direction of the arbitrary element are respectively set to polar coordinates (Px,
0), (Px, π), multiplying the outputs of the elements adjacent in the y-axis direction by a predetermined weight, and calculating an added complementary output. The infrared image sensor according to claim 3, wherein the connected coordinate systems realize two-dimensional arrangement orthogonal to each other. 6. An output of an arbitrary element is represented by S (m, n) (m is x
When the coordinates and n are the y-coordinates, the complementary output F (m, n) multiplied by the weight is expressed as F (m, n) = ｛S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n-1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ or F (m, n) = {S (m, n) * (Px−MOD (m, (QUOTIENT (P
x, Hoffx)) * Hoffx) / Px｝ + ｛S (m, n + 1) * MOD (m, (QUO
TIENT (Px, Hoffx)) * Hoffx) / Px｝ where MOD (a, b) is a remainder obtained by dividing a by b, Q
The infrared image sensor according to claim 5, wherein UOTIENT (a, b) is calculated from a quotient obtained by dividing a by b.