Calculation device, program, anomaly detection method, and calculation method

Abstract

To provide a computing device, program, abnormality detection method, and computation method, which enable computation of an infrared image from a sensor device that offers reduced weight, costs, and installation work to detect abnormalities.SOLUTION: A computing device disclosed herein is for computing an infrared image on the front side of an infrared image sensor using a signal output of the infrared image sensor having a diffractive grating installed on a surface thereof, and is provided with multiple integral kernel numerical tables for integrally converting a first received light intensity function on an x-y coordinate system of the infrared image sensor into a second received light intensity function on a θ-φ coordinate system in a direction of an incident light flux to the infrared image sensor as conversion maps. The computing device comprises an abnormality detection unit configured to compute a required infrared image by switching the conversion maps according to the purpose, and detect an object regarded as a safety threat or detect that ambient temperature is such that a space within a field of view cannot be monitored from the obtained infrared image.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a computing device, a program, an anomaly detection method, and a computing method. [Background technology]

[0002] Conventionally, sensor devices have been disclosed that detect the presence of objects such as people in a predetermined area.

[0003] Patent Document 1 describes a sensor device that uses the pyroelectric effect to detect the presence of a human body as a moving heat source. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2001-118160 [Overview of the project] [Problems that the invention aims to solve]

[0005] However, conventional sensor devices have the drawback of being large and having an undesirable appearance. Furthermore, conventional sensor devices also have the drawbacks of being burdensome to install and making it difficult to flexibly change the installation location.

[0006] The present invention has been made in view of the above, and aims to provide a computing device, program, anomaly detection method, and computing method that calculate infrared images from a sensor device that can be lightweight, low-cost, and require less installation work to detect anomalies. [Means for solving the problem]

[0007] In order to solve the above problems and achieve the object, the present invention is an arithmetic device that calculates an infrared image on the front side of an infrared image sensor using the signal output of the infrared image sensor having a diffraction grating on its surface, and from a first light reception intensity function on the x-y coordinates of the infrared image sensor, a plurality of integral kernel numerical tables for integral conversion to a second light reception intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor are provided as a conversion map, the conversion map is switched according to the application, the necessary infrared image is calculated, and an abnormality detection unit that detects a target object that poses a safety threat from the obtained infrared image or detects that the environmental temperature is such that the field of view cannot be monitored is provided.

Advantages of the Invention

[0008] According to the present invention, there is an effect that an arithmetic device, a program, an abnormality detection method, and an arithmetic method for calculating an infrared image from a sensor device capable of realizing lightweight, low cost, and reduction of installation work and detecting an abnormality can be provided.

Brief Description of the Drawings

[0009] [Figure 1] FIG. 1 is a diagram exemplarily showing a schematic configuration of a sensor system using a sensor device according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the configuration of a sensor control device. [Figure 3] FIG. 3 is a diagram showing a schematic configuration of a sensor device. [Figure 4] FIG. 4 is a cross-sectional view showing the configuration of a sensor device. [Figure 5] FIG. 5 is a diagram showing an example of an infrared blocking pattern of a diffraction grating. [Figure 6] FIG. 6 is a diagram showing the correspondence between a region where infrared rays arrive and a cell of an infrared image sensor that receives infrared rays of the infrared image sensor among regions obtained by dividing the field of view angle. [Figure 7] FIG. 7 is a diagram showing how infrared rays from the field of view region are received by cells of an infrared image sensor. [Figure 8] FIG. 8 is a diagram showing the correspondence between the end portions at the four corners of the divided visual field region and the cells of the infrared image sensor that receive the infrared rays arriving from those end portions. [Figure 9] FIG. 9 is a diagram showing another example of the infrared blocking pattern of the diffraction grating. [Figure 10] FIG. 10 is a diagram showing an example of the diffraction waveform related to the infrared blocking pattern of the diffraction grating. [Figure 11] FIG. 11 is a diagram showing another example of the diffraction waveform related to the infrared blocking pattern of the diffraction grating. [Figure 12] FIG. 12 is a diagram showing an example of enhancing the waveform resolution for an example of the diffraction waveform obtained by diffraction. [Figure 13] FIG. 13 is a diagram showing an example of the configuration of the AC power line. [Figure 14] FIG. 14 is a perspective view showing the structure of an example of the power receiving side power supply coupler and the power transmitting side power supply coupler. [Figure 15] FIG. 15 is a diagram showing the wiring configuration of the power receiving side electromagnetic coil wire of the power receiving side power supply coupler. [Figure 16] FIG. 16 is a cross-sectional view showing an example of the configuration for connecting the sensor device and the AC power line via the power receiving side power supply coupler and the power transmitting side power supply coupler. [Figure 17] FIG. 17 is a cross-sectional view showing another example of the configuration for connecting the sensor device and the AC power line via the power receiving side power supply coupler and the power transmitting side power supply coupler. [Figure 18] FIG. 18 is a cross-sectional view showing yet another example of the configuration for connecting the sensor device and the AC power line via the power receiving side power supply coupler and the power transmitting side power supply coupler. [Figure 19] FIG. 19 is a diagram showing an example of the configuration of the communication line. [Figure 20] FIG. 20 is a flowchart showing the flow of the pre-calculation process in the arithmetic processing of the arithmetic unit. [Figure 21] FIG. 21 is a diagram showing an example of the first sensor data table. [Figure 22]Figure 22 shows an example of a second sensor data table. [Figure 23] Figure 23 shows an example of a two-dimensional response waveform table. [Figure 24] Figure 24 shows an example of an orthonormal basis table. [Figure 25] Figure 25 shows an example of a basis transformation table. [Figure 26] Figure 26 shows an example of an integral kernel table. [Figure 27] Figure 27 is a flowchart showing the process of generating an integral kernel table from an orthonormal basis table. [Figure 28] Figure 28 shows an example of an ingredient table. [Figure 29] Figure 29 is a flowchart showing the flow of real-time calculation processing in the calculation unit. [Figure 30] Figure 30 shows an example of an image table. [Figure 31] Figure 31 is a flowchart showing the process flow for generating an image table from an orthonormal basis table. [Figure 32] Figure 32 is a flowchart showing the flow of anomaly detection and alarm issuance processing. [Figure 33] Figure 33 is a cross-sectional view showing the configuration of a sensor device according to the second embodiment. [Figure 34] Figure 34 shows an example of an infrared light receiving unit. [Figure 35] Figure 35 is a flowchart showing the flow of the detailed image generation process. [Figure 36] Figure 36 shows another example of an infrared light receiving unit. [Figure 37] Figure 37 shows yet another example of an infrared light receiving unit. [Figure 38] Figure 38 shows the results of narrowing down the dominant wavelength by temporally switching the infrared shielding pattern of the diffraction grating. [Figure 39]Figure 39 shows an example of a sensor device and AC power line according to the third embodiment. [Modes for carrying out the invention]

[0010] Embodiments of the arithmetic unit, program, anomaly detection method, and arithmetic method will be described in detail below with reference to the attached drawings. However, this embodiment does not limit the present invention.

[0011] (First Embodiment) Figure 1 is a diagram illustrating the schematic configuration of a sensor system 1 using a sensor device 10 according to the first embodiment. As shown in Figure 1, the sensor system 1 comprises a sensor device 10 installed on the object to be guarded 30 and a sensor control device 20.

[0012] The sensor device 10 detects infrared radiation emitted by a person (heat source) within the detection range of the security target 30 and outputs a signal as image sensor data.

[0013] The sensor control device 20 is a computing device that uses image sensor data output from the sensor device 10 to compute and reconstruct the infrared image on the front side of the sensor. The sensor control device 20 also incorporates a program that uses the infrared image to detect various threats. The sensor control device 20 can be physically mounted anywhere, but since it consumes power for computation, it is connected to the power outlet 21 via a power line / communication line 22.

[0014] The sensor control device 20 is connected to the security system 100 via a network 50, such as a LAN (Local Area Network), and is capable of communicating with each other. The communication method between the sensor control device 20 and the security system 100 may be wireless or wired. The sensor control device 20 transmits, for example, the detection results of various threats to the security system 100.

[0015] Figure 2 is a block diagram showing an example of the configuration of the sensor control device 20. As shown in Figure 2, the sensor control device 20 comprises a control unit 23, a storage unit 24, an external communication unit 25, a calculation unit 26, and a communication unit 27.

[0016] The communication unit 27 controls communication with the sensor device 10 and acquires image sensor data output from the sensor device 10.

[0017] The external communication unit 25 controls communication with the security system 100.

[0018] The control unit 23 is a computer configuration that includes, for example, a CPU (Central Processing Unit), ROM (Read Only Memory) for storing programs, and RAM (Random Access Memory) used as a work area. The control unit 23 implements various functions, such as the CPU executing programs stored in ROM.

[0019] The program executed by the control unit 23 in this embodiment is provided pre-loaded into ROM or the like.

[0020] The program executed by the control unit 23 in this embodiment may be configured to be provided as a file in an installable or executable format, recorded on a computer-readable recording medium such as a CD-ROM, flexible disk (FD), CD-R, or DVD (Digital Versatile Disc).

[0021] Furthermore, the program executed by the control unit 23 of this embodiment may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Alternatively, the program executed by the control unit 23 of this embodiment may be provided or distributed via a network such as the Internet.

[0022] The calculation unit 26 is composed of, for example, an FPGA (Field Programmable Gate Array). The calculation unit 26 uses the image sensor data output from the sensor device 10 to calculate and reconstruct the infrared image on the front side of the sensor.

[0023] The control unit 23 functions as an anomaly detection unit 23a that detects various threats using infrared images restored by the calculation unit 26, such as when the CPU executes a program stored in ROM.

[0024] The memory unit 24 is a memory unit that allows information to be rewritten, such as a Flash ROM. The memory unit 24 functions as a data memory that stores various information used for detecting people (heat sources) within the detection range of the security target 30.

[0025] The calculation unit 26 temporarily stores the image sensor data received via the communication unit 27 into the data memory, and then performs calculations in parallel and sequentially, storing the table values ​​during the calculation into the data memory as they occur.

[0026] The data memory unit 24 functions as a first sensor data table T1, a second sensor data table T2, a two-dimensional response waveform table T3, an orthonormal basis table T4, a basis transformation table T5, an integral kernel table T6, a component table T7, and an image table T8. Details of each table will be described later.

[0027] Next, the sensor device 10 will be described in detail.

[0028] As shown in Figure 1, the sensor device 10 is connected to the power outlet 40 via an AC power line 31, a communication line 32, and a current drive circuit 34. Note that the connection between the sensor device 10 and the AC power line 31 and communication line 32 is not wired, but a short-distance contactless connection. Communication between the power outlet 21 to which the sensor control device 20 is connected and the power outlet 40 to which the sensor device 10 is connected is via the indoor power line 41.

[0029] Here, Figure 3 shows the general configuration of the sensor device 10. As shown in Figure 3, the sensor device 10 is a sheet-shaped sensor device in which an infrared light receiving unit 17 and an integrated circuit 13 are mounted on the surface of a main substrate 14. The infrared light receiving unit 17, as will be described in detail later, has a structure in which a spacer 16 and a diffraction grating 12 are stacked in layers on the surface of an infrared image sensor 11. In other words, the sensor device 10 can be attached via an adhesive sheet or the like. The surface size of the sensor device 10 is approximately a few millimeters square to a few centimeters square. Furthermore, the sensor device 10 does not require a lens, and a thickness of approximately 1 to 2 mm can be achieved.

[0030] Here, Figure 4 is a cross-sectional view showing the configuration of the sensor device 10. Note that wiring is not shown in Figure 4.

[0031] As shown in Figure 4, the sensor device 10 has an infrared light receiving section 17, which has an infrared image sensor 11, a spacer 16, and a diffraction grating 12 stacked in layers, an integrated circuit 13, and a capacitor 15 mounted on top of the main circuit board 14.

[0032] The integrated circuit 13 performs functions such as sensor readout control, sensor setting, and signal transmission. The integrated circuit 13 also includes a power supply circuit.

[0033] The infrared image sensor 11 has a dot accuracy of approximately 1 to 10 μm.

[0034] The diffraction grating 12 is an infrared blocking pattern forming layer composed of a reflective liquid crystal dot matrix. The diffraction grating 12 can form any diffraction grating by using an external signal to make any group of dots an infrared reflective surface to block infrared rays, while allowing infrared rays to pass through other groups of dots, and this can be changed in units of a few milliseconds to tens of milliseconds. The diffraction grating 12 has a thickness of approximately 5 μm.

[0035] A spacer 16, which is an infrared-transmitting layer, is provided between the infrared image sensor 11 and the diffraction grating 12. The spacer 16 has a thickness of approximately 10 to 100 μm. More specifically, the distance between the infrared image sensor 11 and the diffraction grating 12 is the same as, or several times thicker than, the main wavelength of infrared radiation emitted by the person (heat source) being detected (approximately 10 μm at around 36°C). The sensor device 10, consisting of the infrared image sensor 11, the diffraction grating 12, and the spacer 16, constitutes an infrared light receiving unit 17 that receives infrared radiation within the field of view.

[0036] As shown in Figure 4, the sensor device 10 includes, at the bottom of the main board 14, a plurality of signal receiving and transmitting couplers 18a, which are receiving signal transmitting and receiving circuits that transmit and receive signals by converting signals into AC magnetic field lines and emitting them, or receiving AC magnetic field lines and converting them into signals, and a plurality of receiving-side power supply couplers 18b, which are receiving circuits that receive AC magnetic field lines from an external source and convert them into power.

[0037] Furthermore, the sensor device 10 is equipped with a non-conductive adhesive sheet 19 on the outer surfaces of the signal transmitting / receiving coupler 18a and the power receiving side power supply coupler 18b. As will be described in detail later, the sensor device 10 is installed so that the signal transmitting / receiving coupler 18a and the power receiving side power supply coupler 18b are in contact with the adhesive sheet 19 or via a non-conductive wall, respectively, and are facing the communication line 32 and the AC power line 31, and are electromagnetically coupled via AC magnetic field lines. The communication line 32 is a signal transmitting / receiving circuit that transmits signals by converting them into AC magnetic field lines or receives AC magnetic field lines and converts them into signals to transmit and receive signals with the signal transmitting / receiving coupler 18a. The power transmitting side power supply coupler 38b is a power transmission circuit that transmits power by converting it into AC magnetic field lines and has a structure similar to that of the power receiving side power supply coupler 18b (see Figure 16, etc.).

[0038] Here, we will describe the diffraction grating 12 in detail.

[0039] Figure 5 shows an example of an infrared blocking pattern of the diffraction grating 12. The shaded area of ​​the diffraction grating 12 is the infrared blocking surface, and the white area is an opening through which infrared light is transmitted. In the example shown in Figure 5, there is one infrared blocking pattern forming layer. Figure 5(a) is a top view of the diffraction grating 12 constituting the infrared light receiving unit 17, and Figure 5(b) is an example of a cross-sectional view of the diffraction grating 12, spacer 16, and infrared image sensor 11 constituting the infrared light receiving unit 17 in AA of Figure 5(a).

[0040] In the example shown in Figure 5(b), the size of the aperture (white) of the diffraction grating 12 is approximately 100 μm (a size that causes almost no diffraction at a wavelength of 10 μm in infrared light). The spacing between the apertures (white) is approximately 80 μm. In Figure 5(a), the middle of the aperture arrangement (indicated by "···" in the figure) is omitted.

[0041] In this embodiment, the field of view is divided into approximately 9x9 regions, and the infrared intensity of each region is imaged. Human movement is detected by detecting movement in regions where an infrared intensity above a certain level is detected. Therefore, when the size of the infrared image sensor 11 cells is 10 μm, an aperture spacing of approximately 10 μm x 8 = 80 μm can be used to arrange the infrared image sensor 11 cells corresponding to a 9x9 resolution.

[0042] Furthermore, in the example shown in Figure 5(b), if the thickness of the infrared-transmitting spacer 16 is 10 μm, the range of the infrared image sensor 11 cell that receives obliquely incident light beams with an incident angle of 60° or less is within approximately 10 μm × √3 = 18 μm to the side of the edge of the aperture. Therefore, if the aperture spacing is approximately 80 μm, the two oblique light beams with incident angles of 60° and -60° will not interfere with each other on the cell of the infrared image sensor 11. Note that the amount of light beam at an incident angle of 60° decreases to cos60° = 0.5, so the amount of light beam at incident angles greater than that is considered to be an error and is ignored.

[0043] If the infrared image sensor 11 is 2 mm in size and the aperture size is as described above, then 2 mm / (100 μm + 80 μm) = 11.1 vertically, and similarly horizontally, approximately 10 x 10 = 100 apertures can be arranged within the sensor device 10.

[0044] Here, Figure 6 shows the correspondence between the region where infrared light arrives within the divided field of view (hereinafter referred to as the "field of view region") and the cells of the infrared image sensor 11 that receive the infrared light. Figure 6(a) shows the region (shaded area) within the divided field of view region where infrared light arrives when the incident angle θ is -5° to +5° to the infrared light receiving unit 17. Similarly, Figures 6(c) and 6(e) show the same region when the incident angle θ is +15° to +30° and when the incident angle θ is +45° to +60°, respectively. Figure 6(b) is a cross-sectional view of the infrared light receiving unit 17 in AA of Figure 5(a), showing how infrared light with an incident angle θ of -5° to +5° is incident around one of the multiple apertures of the diffraction grating 12. Similarly, Figures 6(d) and 6(f) show how infrared light is incident when the incident angle θ is +15° to +30° and +45° to +60°, respectively. In addition, in Figures 6(b), (d), and (f), the bending of light occurs due to the difference in refractive index between the air and the infrared blocking pattern forming layer or infrared transmitting layer, but this is not shown.

[0045] Here, Figures 7(a), (b), and (c) show how infrared light from the field of view shown in Figures 6(a), (c), and (e), respectively, is received by the cells of the infrared image sensor 11. In Figure 7, only the cells of the infrared image sensor 11 around one of the approximately 100 apertures of the diffraction grating 12 are shown. In Figure 7(a), the incident angle θ of infrared light from the field of view is -5° to +5°, in Figure 7(b), the incident angle θ of infrared light from the field of view is +15° to +30°, and in Figure 7(c), the incident angle θ of infrared light from the field of view is +45° to +60°.

[0046] In Figure 7, the sum of the output values ​​of the shaded cells of the infrared image sensor 11 corresponds to the amount of infrared radiation in the field of view. This point will be explained in detail below.

[0047] Figure 8 shows the correspondence between the four corners of the divided field of view and the cells of the infrared image sensor 11 that receive infrared light arriving from those corners. Figure 8(a) shows the four corners of the divided field of view, i.e., the region where infrared light from that region is incident on the infrared light receiving unit 17 at an incident angle φ of -60° to -45° and an incident angle θ of -60° to -45°, the region where the incident angle φ of -60° to -45° and an incident angle θ of +45° to +60°, the region where the incident angle φ of +45° to +60° and an incident angle θ of +45° to +60°, and the region where the incident angle φ of +45° to +60° and an incident angle θ of -60° to -45°. Figures 8(b) and 8(c) show the correspondence between each of the four corner field of view regions and the cells of the infrared image sensor 11 that receive infrared light. In Figures 8(b) and 8(c), only the vicinity of one of the approximately 100 apertures of the diffraction grating 12 is shown.

[0048] Infrared light from all incidence angles mixes and reaches the cells of the infrared image sensor 11. However, for the four endpoints of the field of view shown in Figure 8(a), only infrared light from that incidence angle reaches the corresponding cells of the infrared image sensor 11 shown in Figure 8(b) (the four types of shading indicate correspondence). Therefore, by using the infrared detection values ​​at the four endpoints of the cells of the infrared image sensor 11 shown in Figure 8(b), and sequentially subtracting them from the infrared detection values ​​of adjacent cells where infrared light from different incidence angles is detected, the infrared detection values ​​from all specific incidence angles can be calculated. For example, suppose that only infrared light (a) from incidence angles φ of -60° to -45° and incidence angles θ of -60° to -45° is incident on the upper right endpoint of the cell of the infrared image sensor 11 shown in Figure 8(b). Furthermore, at the point adjacent to the upper right endpoint of the cell of the infrared image sensor 11 shown in Figure 8(c), infrared radiation (a) from an incident angle φ of -60° to -45° and an incident angle θ of -60° to -45°, and infrared radiation (b) from an incident angle φ of -45° to -30° and an incident angle θ of -60° to -45° are incident together. However, by subtracting the infrared detection value of a cell where only infrared radiation (a) is incident from the infrared detection value of that cell, the infrared detection value due to infrared radiation (b) alone can be calculated.

[0049] Here, Figure 9 shows another example of the infrared blocking pattern of the diffraction grating 12. In the example shown in Figure 9, there is one infrared blocking pattern forming layer. Figure 9 is a cross-sectional view of the diffraction grating 12, spacer 16, and infrared image sensor 11 that constitute the infrared light receiving section 17 in AA of Figure 5(a). The diffraction grating 12 shown in Figure 9 has a thickness of about 10 μm, the spacer 16 has a thickness of about 50 μm, and the dot accuracy of the infrared image sensor 11 is about 3 μm.

[0050] As shown in Figure 9, the size of the aperture (white) of the diffraction grating 12 is approximately 10 μm (the size at which infrared light with a wavelength of 10 μm spreads due to diffraction and enters the cell of the infrared image sensor 11). The spacing between the apertures (white) is approximately 90 μm. In the example shown in Figure 9, infrared light with a wavelength of 10 μm from adjacent apertures interferes, generating a periodic pattern on the sensor surface of the infrared image sensor 11.

[0051] If the infrared image sensor 11 is 2 mm in size and the aperture size is as described above, then 2 mm / (10 μm + 90 μm) = 20 vertically, and similarly horizontally, approximately 20 x 20 = 400 apertures can be arranged within the sensor device 10.

[0052] As described above, since the aperture (white) is about 10 μm in size and the infrared blocking pattern formation layer is about 10 μm thick, incident light beams (the zero-order diffraction component) with an incident angle of 45° or more do not reach the sensor surface of the infrared image sensor 11. If the infrared image sensor 11 is 2 mm in size and has a dot accuracy of 1 μm, it has the potential for a resolution of 2000 × 2000 if a lens is used, but as will be explained later, Figure 9 assumes a resolution of about 10 × 10 or less.

[0053] Here, Figure 10 shows an example of diffraction waveforms applied to the infrared blocking pattern of the diffraction grating 12. Figure 10(a) is a cross-sectional view of the diffraction grating 12, spacer 16, and infrared image sensor 11 that constitute the infrared light receiving section 17 in AA of Figure 5(a), Figure 10(b) is an example of diffraction waveforms from a specific field of view (θ=0°), and Figure 10(c) is an example of diffraction waveforms from a specific field of view (θ=20°). The diffraction grating 12 shown in Figure 10(a) has an aperture (white) size of about 10 μm, an aperture (white) spacing of about 90 μm, and a thickness of about 10 μm, the spacer 16 has a thickness of about 50 μm, and the dot accuracy of the infrared image sensor 11 is about 3 μm.

[0054] Figure 11 shows another example of the diffraction waveform applied to the infrared blocking pattern of the diffraction grating 12. Figure 11(a) is a cross-sectional view of the infrared light receiving section 17 in AA of Figure 5(a), Figure 11(b) is an example of the diffraction waveform from a specific field of view (θ=0°), and Figure 11(c) is an example of the diffraction waveform from a specific field of view (θ=1°). The diffraction grating 12 shown in Figure 11(a) has an aperture (white) size of about 10 μm, an aperture (white) spacing of about 90 μm, a thickness of about 10 μm, a spacer 16 with a thickness of about 50 μm, and a dot accuracy of about 3 μm for the infrared image sensor 11. Note that the origin of the horizontal coordinate x of the diffraction waveforms in Figures 10(b), 10(c), 11(b), and 11(c) is directly below the aperture.

[0055] As shown in Figures 10 and 11, the detection resolution of the waveform is low because the number of cells in the infrared image sensor 11 is small compared to the diffraction waveform obtained by diffraction. However, the number of cells in the infrared image sensor 11 directly below the period (aperture size + spacing) of the diffraction grating 12 can sometimes improve the waveform resolution when the period of the diffraction waveform and the sampling interval are not integer multiples.

[0056] Here, Figure 12 shows an example of improving the waveform resolution of a diffraction waveform obtained by diffraction. Figure 12 is an example of a diffraction waveform from a specific field of view (θ=0°).

[0057] When the infrared blocking pattern of the diffraction grating 12 is periodic, the distance from the diffraction grating 12 to the target field of view is extremely large compared to the distance from the diffraction grating 12 to the sensor surface of the infrared image sensor 11. Therefore, the light beam from one point in the target field of view is incident on all apertures at approximately the same angle of incidence. Consequently, the resulting diffraction waveform is also periodic.

[0058] The arrows in Figure 12 correspond to the spacing between cells of the finite infrared image sensor 11 and indicate the sampling interval for detecting the light intensity of the diffraction waveform. As shown in Figure 12, all sampling values ​​with different phases for the repeating waveform are used.

[0059] As shown in Figure 12, because the number of cells in the infrared image sensor 11 is small compared to the diffraction waveform obtained by diffraction, the detection resolution of the waveform becomes low when the period of the diffraction waveform and the sampling interval are exactly integer multiples.

[0060] On the other hand, if, for example, 10 cells of the infrared image sensor 11 correspond to one period of the diffraction grating 12, then the signal-to-noise ratio can be increased by adding the corresponding sampling values ​​for the entire period. In other words, if the number of cells of the infrared image sensor 11 directly below the period of the diffraction grating 12 (aperture size + spacing) is not an integer multiple of the period of the diffraction waveform and the sampling interval, the waveform detection resolution can be increased by sampling and combining different phases of the repeating waveform, even if the sampling period does not become dramatically narrower.

[0061] For example, if 38 cells of the infrared image sensor 11 correspond to 5 periods of the diffraction grating 12, that is, if there are an average of 7.6 cells per period of the diffraction grating 12, then because 5 and 38 are relatively prime, when adding up the samples for all periods, corresponding sampling points are obtained in units of 5 periods. As a result, this corresponds to 38 samples per repeating waveform period, but the signal-to-noise ratio will be slightly inferior.

[0062] Next, we will explain the AC power line 31 to which the sensor device 10 is attached.

[0063] Figure 13 shows an example of the configuration of the AC power line 31. As shown in Figure 13, the AC power line 31 is connected to the indoor power line 41 via a current drive circuit 34. The AC power line 31 is installed on the front side of a non-conductive building material that constitutes a wall or partition panel (screen) of a building via a sealing tape-like adhesive tape 35. The AC power line 31 can also be installed on the back or inside of a thin insulating wall material. That is, the building material has the power transmission circuit inside the building material or on a surface that is not exposed. As shown in Figure 13, the AC power line 31 is equipped with power transmission side electromagnetic coil wires 71 (71a~71c) that constitute the power transmission side power supply coupler 38b. Note that the shape of the power transmission side electromagnetic coil wire 71 in Figure 13 is a simplified representation. More specifically, the power transmission side electromagnetic coil wires 71a~71c are a pair of AC current lines drawn in parallel from an intermediate point of a pair of AC power lines 31 and a power supply circuit connected to its endpoints. The sensor device 10 is attached to the required position among the multiple transmitting electromagnetic coil wires 71a to 71c. In the example shown in Figure 13, the sensor device 10 receives power from the AC power line 31 via the AC magnetic field lines by attaching the receiving power supply coupler 18b to the transmitting power supply coupler 38b, which has the uppermost transmitting electromagnetic coil wire 71a, via a non-conductive adhesive sheet 19 (not shown). Also, as shown in Figure 13, high inductance seals 36 (an example of a high inductance element) are attached to the empty transmitting electromagnetic coil wires 71b and 71c to which the receiving power supply coupler 18b of the sensor device 10 is not attached. The high-inductance seal 36 has more turns (higher inductance) than the receiving-side power supply coupler 18b, and as a result acts to increase the impedance of the transmitting-side electromagnetic coil wires 71b and 71c, exhibiting a high impedance value. Therefore, current is less likely to flow through the transmitting-side electromagnetic coil wires 71b and 71c, and leakage of AC magnetic field lines can be suppressed. Alternatively, instead of the high-inductance seal 36, a switch element may be interposed between the transmitting-side electromagnetic coil wire 71 and the AC current line pair. The switch element connected to the transmitting-side electromagnetic coil wire 71a used for power supply to the sensor device 10 may be turned on, and the switch elements connected to the unused transmitting-side electromagnetic coil wires 71b and 71c may be turned off to suppress leakage of AC magnetic field lines.

[0064] Next, the structure of the receiving-side power supply coupler 18b and the transmitting-side power supply coupler 38b will be described.

[0065] Figure 14 is a perspective view showing the structure of an example of a receiving-side power supply coupler 18b and a transmitting-side power supply coupler 38b. As shown in Figure 14, the receiving-side power supply coupler 18b of the sensor device 10 has a configuration in which a high-permeability sheet (ferromagnetic sheet) 63 is bonded from both above and below to a plurality of high-permeability blocks 62, each containing a receiving-side electromagnetic coil wire 61 as a closed-curve type power supply wiring. The high-permeability block 62 is, for example, an iron core located at the center of the coil. The high-permeability sheet 63 is, for example, an iron metal foil. The high-permeability sheets 63 are installed intersectingly between the receiving-side electromagnetic coil wires 61 (see also Figure 16, etc.).

[0066] As shown in Figure 14, the power transmission-side power supply coupler 38b and the power reception-side power supply coupler 18b have similar structures (see also Figure 16, etc.). For example, the power transmission-side power supply coupler 38b has a configuration in which a high-permeability sheet (ferromagnetic sheet) 73 is bonded from both above and below to a plurality of high-permeability blocks 72, each containing a power transmission-side electromagnetic coil wire 71 as a closed-curve type power supply wiring. The high-permeability sheet 73 is, for example, iron metal foil. The high-permeability sheet 73 is installed intersectingly between the power transmission-side electromagnetic coil wires 71 (see also Figure 16, etc.).

[0067] Figure 15 shows the wiring configuration of the receiving side electromagnetic coil wire 61 of the receiving side power supply coupler 18b. Figure 15(a) shows the relationship between the receiving side electromagnetic coil wire 61 and the switch 64, and Figure 15(b) shows the configuration from the switch 64 to the capacitor 15 or integrated circuit 13. As shown in Figure 15(b), current summing circuits (transistors) 65 and 66 are used to add the alternating currents from the receiving side electromagnetic coil wires 61, which are multiple closed-curve type power supply wirings, while controlling the positive and negative polarity of each, thereby converting the alternating current into a ripple current with less fluctuation.

[0068] Here, Figure 16 is a cross-sectional view showing an example configuration in which the sensor device 10 and the AC power line 31 are connected via a receiving-side power supply coupler 18b and a transmitting-side power supply coupler 38b. The example shown in Figure 16 shows a case where the tape-shaped AC power line 31 is attached to the surface of a wall. As shown in Figure 16, the receiving-side power supply coupler 18b and the transmitting-side power supply coupler 38b have similar structures, and between them, AC magnetic field lines (frequency of approximately 50-100 kHz) are generated by electromagnetic induction interaction, enabling wireless power supply. In Figure 16, the AC power line 31 is covered with an insulating protective member 37.

[0069] Figure 17 is a cross-sectional view showing another configuration example in which the sensor device 10 and the AC power line 31 are connected via a receiving-side power supply coupler 18b and a transmitting-side power supply coupler 38b. The example shown in Figure 17 shows a case where the AC power line 31 is attached to the back of a wall. As shown in Figure 17, AC magnetic field lines (frequency of approximately 50-100 kHz) are generated between the receiving-side power supply coupler 18b and the transmitting-side power supply coupler 38b due to electromagnetic induction interaction, enabling wireless power supply. The high-permeability sheet 63 constituting the receiving-side power supply coupler 18b is designed to suppress leakage of AC magnetic field lines even if the position of the receiving-side power supply coupler 18b is slightly shifted. In Figure 17, the AC power line 31 is covered with an insulating protective member 37.

[0070] Figure 18 is a cross-sectional view showing yet another configuration example in which the sensor device 10 and the AC power line 31 are connected via a receiving-side power supply coupler 18b and a transmitting-side power supply coupler 38b. The example shown in Figure 18 shows a case where the AC power line 31 is attached to the back of a wall. As shown in Figure 18, AC magnetic field lines (frequency of approximately 50-100 kHz) are generated between the receiving-side power supply coupler 18b and the transmitting-side power supply coupler 38b by electromagnetic induction interaction, enabling wireless power supply. The high-permeability sheet 63 constituting the receiving-side power supply coupler 18b is designed to suppress leakage of AC magnetic field lines even if the position of the receiving-side power supply coupler 18b is slightly shifted. In Figure 18, the AC power line 31 is covered with an insulating protective member 37.

[0071] Next, the connection between the sensor device 10 and the communication line 32 will be explained.

[0072] Here, Figure 19 shows an example of the configuration of the communication line 32. As shown in Figure 19, the communication line 32 is fitted with an evanescent communication line seal. The evanescent communication line seal is a signal line comprising a dielectric 32a and a signal conductor mesh 32b. The dielectric 32a and the signal conductor mesh 32b are sandwiched between a surface insulator 32c and a ground conductor wire 32d and attached with an adhesive 32e. The evanescent communication line seal transmits image sensor data output from the sensor device 10 using evanescent waves, for example, by the method described in Japanese Patent Application Publication No. 2007-150652. The signal transmitting / receiving coupler 18a of the sensor device 10 is attached to the surface insulator 32c side of the communication line 32.

[0073] Next, we will describe the calculation process in the calculation unit 26 of the sensor control device 20 that calculates the infrared image on the front side of the sensor using the signal output of the sensor device 10. In this embodiment, we will describe a method for generating a target field of view image from a lensless infrared image sensor 11.

[0074] In general, the system includes multiple integral kernel numerical tables as conversion maps that perform an integral transformation from the signal output of the sensor device 10 (a first light intensity function on the xy coordinates of the infrared image sensor 11) to an infrared image (a second light intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor 11). The system switches between these conversion maps according to the application to calculate the required infrared image.

[0075] Here, Figure 20 is a flowchart showing the flow of the pre-calculation process in the calculation processing of the calculation unit 26. As shown in Figure 20, the calculation unit 26 starts the pre-calculation.

[0076] First, the calculation unit 26 receives the output signal (sensor data) on the two-dimensional coordinate (x,y) of the infrared image sensor 11 from only a specific incident angle (step S1) and stores it in the first sensor data table T1.

[0077] Here, Figure 21 shows an example of the first sensor data table T1. As shown in Figure 21, the first sensor data table T1 is a table that stores the output values ​​of the two-dimensional array infrared image sensor 11 immediately after sampling. For example, in the case of an infrared image sensor 11 with 1000 x 800 cells (pixels), M=1000 and N=800.

[0078] Next, the calculation unit 26 reads sensor data for each diffraction grating repetition period from the first sensor data table T1, interpolates and generates the response waveform of the sensor output in units of diffraction grating repetition (step S2), and writes it to the second sensor data table T2.

[0079] Here, Figure 22 shows an example of the second sensor data table T2. As shown in Figure 22, the second sensor data table T2 is a table that stores data that has been interpolated based on the output values ​​of the two-dimensional array image sensor (first sensor data table T1) so that the data is for each diffraction grating period. For example, if there are 10 image sensor cells in one diffraction grating period (sampling example in Figure 11(c)), then m=10 and n=10, and if there are 38 image sensor cells in five diffraction grating periods (sampling example in Figure 11(d)), then m=38 and n=38.

[0080] Next, the calculation unit 26 outputs the data stored in the second sensor data table T2 to the two-dimensional response waveform table T3 for each incident angle (step S3).

[0081] Here, Figure 23 shows an example of a two-dimensional response waveform table T3. As shown in Figure 23, the two-dimensional response waveform table T3 is the same size as the second sensor data table T2, and stores the sensor surface waveform (second sensor data table T2) for infrared incidence only in the specified direction θ on the horizontal axis and φ on the vertical axis of the field of view. If the field of view is divided into s divisions in the horizontal axis θ direction and t divisions in the vertical axis φ direction, then there are s × t tables in the two-dimensional response waveform table T3. In the case where there are 10 image sensor cells in one period of the diffraction grating (sampling example in Figure 11(c)), the two-dimensional response waveform table T3 has m=10 and n=10, and in the case where there are 38 image sensor cells in five periods of the diffraction grating (sampling example in Figure 11(d)), it has m=38 and n=38.

[0082] The calculation unit 26 repeats the processing in steps S1 to S3 (No. in step S4) while changing the incident angle by the number of region resolutions required for the image resolution.

[0083] If the processing in steps S1 to S3 is repeated for all image resolution regions (Yes in step S4), the calculation unit 26 calculates an orthonormal basis from each response in the 2D response waveform table T3 based on the Gran-Schmitt orthogonalization method (step S5).

[0084] Next, the calculation unit 26 outputs the calculation results to the orthonormal basis table T4 and the basis transformation table T5 (step S6).

[0085] Here, FIG. 24 is a diagram showing an example of the orthonormal basis table T4. As shown in FIG. 24, the orthonormal basis table T4 has the same size as the second sensor data table T2, and is a table showing the waveforms of the orthonormal basis obtained through the Gram-Schmitt orthogonalization method from the two-dimensional response waveform table, and there are as many orthonormal bases as the number of orthonormal bases. If the orthonormal basis table T4 is divided into s in the horizontal axis θ direction of the viewing angle and t in the vertical axis φ direction, there are s×t tables. When there are 10 image sensor cells per diffraction grating period (sampling example in FIG. 11(c)) in the orthonormal basis table T4 shown in FIG. 24, m = 10 and n = 10, and when there are 38 image sensor cells per 5 diffraction grating periods (sampling example in FIG. 11(d)), m = 38 and n = 38. The orthonormal basis table T4 shown in FIG. 24 displays the orthonormal basis B normalized by the Gram-Schmitt method, and the following relationship holds (for the display of β, refer to FIG. 25). B st mn =β st s’t’ ·R s’t’ mn

[0086] Here, FIG. 25 is a diagram showing an example of the basis conversion table T5. As shown in FIG. 25, the basis conversion table T5 is a table that stores the expansion coefficients obtained by expanding each waveform of the two-dimensional response table (refer to FIG. 23) with the waveforms of the orthonormal basis table T4 (refer to FIG. 24), and is used for the mutual conversion of the above two types of bases. β is the conversion coefficient between the two. B st mn =β st s’t’ ·R s’t’ mn

[0087] Next, the arithmetic unit 26 generates the integration kernel table T6 from the orthonormal basis table T4 (step S7) and ends the pre-calculation. The integration kernel table T6 is used to calculate the viewing intensity on the (θ, φ) coordinates from the sensor surface waveform within the lattice unit region (x, y).

[0088] Here, Figure 26 shows an example of the integral kernel table T6. As shown in Figure 26, the integral kernel table T6 is the transformation coefficient obtained from the two tables in Figures 23 and 25. The integral kernel table T6 is the same size as the second sensor data table T2, and there are as many tables showing the function shapes of the orthonormal basis as there are orthonormal bases. If the integral kernel table T6 is divided into s divisions in the θ direction and t divisions in the φ direction, then there are s × t tables. σ -1 mn ·β -1 st s’t’ This corresponds to the integral kernel table T6.

[0089] Here, we will explain how to generate the integral kernel table T6 from the orthonormal basis table T4.

[0090] Figure 27 is a flowchart showing the process flow for generating the integral kernel table T6 from the orthonormal basis table T4. As shown in Figure 27, the calculation unit 26 decomposes the response waveform (on x,y coordinates) of the sensor output for a single signal from a specific field of view direction (θ,φ) into components for each basis of the orthonormal basis table T4 and outputs them to the component table T7 (step S11).

[0091] Here, Figure 28 shows an example of the component table T7. As shown in Figure 28, the component table T7 is a table that decomposes the data from the second sensor data table T2 into components for each orthonormal basis and shows the magnitude of each component. The horizontal axis of the table corresponds to the horizontal axis θ of the field of view, and the vertical axis of the table corresponds to the vertical axis φ of the field of view. If the number of orthonormal basis bases is s × t, it becomes a single table of size s × t, and the component for the s-th horizontal and t-th vertical basis is stored in the column s-1 horizontally and t-1 vertically in the lower right of the component table T7. The waveform D' on the sensor surface can be expanded using the orthonormal basis B, and its expansion coefficient is α. D' mn =α s’t’ ·B s’t’ mn

[0092] Next, the arithmetic unit 26 calculates an inverse basis for each orthonormal basis, taking the x and y coordinate components as inputs and the θ and φ coordinate components as outputs (matrix operation between the table in Figure 23 and the table in Figure 25) (step S12).

[0093] Next, the calculation unit 26 outputs the calculation results as an inverse basis to the inverse basis table (integral kernel table T6) (step S13).

[0094] Next, the calculation unit 26 repeats the processing in steps S11 to S13 while changing the specific field of view direction (θ, φ) (No. in step S14).

[0095] Then, when all calculations are completed (Yes in step S14), the arithmetic unit 26 terminates processing.

[0096] Next, Figure 29 is a flowchart showing the flow of real-time calculation processing in the calculation processing of the calculation unit 26. As shown in Figure 29, the calculation unit 26 receives the output signal (sensor data) on the two-dimensional coordinate (x,y) of the infrared image sensor 11 at the motion image sampling timing (step S21) and stores it in the first sensor data table T1.

[0097] Next, the calculation unit 26 reads sensor data for each diffraction grating repetition period from the first sensor data table T1, generates an interpolated response waveform of the sensor output in units of diffraction grating repetition (step S22), and writes it to the second sensor data table T2.

[0098] Next, the calculation unit 26 calculates the field of view intensity in a specific angle of view direction (θ,φ) coordinate using the integral kernel table T6 (step S23).

[0099] Next, the calculation unit 26 outputs the calculation result to the image table T8 (step S24).

[0100] Here, Figure 30 shows an example of an image table T8. As shown in Figure 30, the image table T8 stores image data obtained by adding up all the image component waveforms, which are obtained by multiplying the values ​​for θ and φ in each of the s × t integral kernel tables T6 by the corresponding values ​​in the component table T7.

[0101] The calculation unit 26 repeats the processing in steps S23 to S24 while changing the specific field of view direction (θ, φ) (No. in step S25).

[0102] When the calculation unit 26 has finished all calculations (Yes in step S25), it repeats the same calculations for steps S21 to S25 at the next video sampling timing (No in step S26).

[0103] Then, when all calculations are completed (Yes in step S26), the arithmetic unit 26 terminates processing.

[0104] Here, we will explain how to generate an image table T8 from an orthonormal basis table T4.

[0105] Figure 31 is a flowchart showing the process flow for generating an image table T8 from an orthonormal basis table T4. As shown in Figure 31, the calculation unit 26 decomposes the response waveform (on x,y coordinates) of the sensor output of the infrared image sensor 11 into components for each basis of the orthonormal basis table T4 and outputs them to the component table T7 (step S31).

[0106] Next, the calculation unit 26 calculates the image by matrix operations on the component table T7 and the integral kernel table T6 (step S32). The following formula is an example of a calculation formula. I st =σ -1 mn ·β -1 st s’t’ ·α -1 s’t’ ·D' mn D': Intensity waveform on the sensor surface (Figure 22) I st Image (Figure 30) σ: 2D response (Figure 23) α: Expansion coefficients based on an orthonormal basis (Figure 28) β: Transformation coefficient of the two types of basis (Figure 25) σ -1 mn ·β -1 st s’t’ This corresponds to the integral kernel table T6 (Figure 26). The subscripts s and t correspond to the θ and φ directions in the field of view. The subscripts m and n correspond to the x and y directions on the sensor surface.

[0107] Next, the calculation unit 26 outputs the calculation result to the image table T8 (step S33).

[0108] Next, the anomaly detection and alarm issuance by the anomaly detection unit 23a will be described. The anomaly detection unit 23a detects an object that poses a safety threat based on the image obtained above, or detects that the ambient temperature is such that monitoring within the field of view is not possible, and outputs a warning.

[0109] Here, Figure 32 is a flowchart showing the flow of anomaly detection and alarm issuance processing. As shown in Figure 32, the anomaly detection unit 23a first monitors the intensity change using a low-resolution image (step S101).

[0110] Next, the anomaly detection unit 23a detects an increase in intensity in the majority of pixels (step S102).

[0111] If an increase in intensity is detected in the majority of pixels (Yes in step S102), the anomaly detection unit 23a issues an alarm indicating that detection is impossible (step S103), and the process returns to step S101.

[0112] If an increase in intensity is not detected in the majority of pixels (No. in step S102), the anomaly detection unit 23a detects an increase in intensity in a specific pixel or a shift in intensity to an adjacent pixel (step S104).

[0113] If an increase in the intensity of a specific pixel or a shift in intensity to an adjacent pixel is detected (Yes in step S104), the anomaly detection unit 23a issues a primary alarm (step S105).

[0114] On the other hand, if no increase in intensity of a specific pixel or shift in intensity to an adjacent pixel is detected (No. in step S104), the anomaly detection unit 23a returns to step S101.

[0115] Next, the anomaly detection unit 23a confirms the object of change using a high-resolution image centered on the change point (step S106), and determines that the object of change is an intruder (step S107).

[0116] If the device is not identified as an intruder (No. in step S107), the anomaly detection unit 23a returns to step S101.

[0117] On the other hand, if the change is determined to be an intruder (Yes in step S107), the anomaly detection unit 23a issues a secondary alarm (alarm cause) (step S108), returns to step S101, and repeats the same calculation at the video sampling timing (No in step S109).

[0118] Then, when all calculations are completed (Yes in step S109), the anomaly detection unit 23a terminates the process.

[0119] As described above, the anomaly detection unit 23a detects the presence of a moving object within the field of view from the time change of the calculated infrared image, and detects the movement of the target object by detecting a high-resolution image only in the angular direction in which the object is located.

[0120] As described above, this embodiment eliminates the need for a lens, enabling the realization of an infrared human presence sensor that is as thin as a felt seal (approximately 1-2 mm), lightweight, and low-cost. This reduces installation work and allows for use in fire detection and monitoring people within a target field of view. Furthermore, this embodiment enables efficient anomaly detection based on the size, shape, distance, and direction of the target object.

[0121] Furthermore, according to this embodiment, a reflective liquid crystal dot matrix can be used to form any diffraction pattern by making any group of dots an infrared reflecting surface by an external signal, and an arbitrary diffraction pattern can be formed by a diffraction grating that can change in units of several milliseconds to tens of milliseconds. By sequentially switching between a first diffraction pattern with aperture size a, a second diffraction pattern with aperture size b, ..., the Nth diffraction pattern with aperture size z, and by simultaneously switching corresponding conversion maps (multiple integral kernel numerical tables that perform an integral transformation from the signal output (first received intensity function on the xy coordinates of the infrared image sensor) to the infrared image (second received intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor)), it is possible to discriminate infrared light of a specific wavelength and calculate an infrared image.

[0122] Furthermore, using a reflective liquid crystal dot matrix, which can form arbitrary diffraction patterns by making any group of dots an infrared reflector using an external signal, and which can change in units of several milliseconds to tens of milliseconds, a series of time-series diffraction patterns using a pseudo-random pattern, which is a pseudo-orthogonal sequence adjusted for the purpose of detecting time variations in the image, and a corresponding transformation map (multiple integral kernel numerical tables that perform an integral transformation from the signal output (first received intensity function on the xy coordinates of the infrared image sensor) to the infrared image (second received intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor)) can be used to detect the presence of motion of an object.

[0123] (Second Embodiment) Next, a second embodiment will be described.

[0124] The second embodiment differs from the first embodiment in that the diffraction grating has a two-layer structure (multilayer structure). In the following description of the second embodiment, the parts that are the same as those in the first embodiment will be omitted, and the parts that differ from the first embodiment will be described.

[0125] Figure 33 is a cross-sectional view showing the configuration of the sensor device 10 according to the second embodiment. Note that wiring is not shown in Figure 33. As shown in Figure 33, the sensor device 10 differs from the sensor device 10 shown in Figure 4 in that the diffraction grating has a two-layer structure (multilayer structure). In this embodiment, the diffraction grating is described as having a two-layer structure, but three or more layers are also possible.

[0126] As shown in Figure 33, the sensor device 10 comprises a diffraction grating 12a which is a first infrared blocking pattern forming layer, an infrared first transmission layer 16a, a diffraction grating 12b which is a second infrared blocking pattern forming layer, and an infrared second transmission layer 16b. The diffraction grating 12a has a thickness of about 5 μm and a dot accuracy of about 1 to 10 μm. The infrared first transmission layer 16a has a thickness of about 10 to 100 μm. The diffraction grating 12b has a thickness of about 5 μm and a dot accuracy of about 1 to 10 μm. The infrared second transmission layer 16b has a thickness of about 10 to 100 μm. The sensor device 10, consisting of an infrared image sensor 11, diffraction gratings 12a and 12b, and spacers 16a and 16b, constitutes an infrared light receiving unit 17 that receives infrared light within the field of view.

[0127] Figure 34 shows an example of an infrared light receiving unit 17. As shown in Figure 34, the infrared light receiving unit 17 limits the target field of view with a diffraction grating 12a, which is the first infrared blocking pattern forming layer, and diffracts with a diffraction grating 12b, which is the second infrared blocking pattern forming layer. If an aperture of several times the wavelength is secured in order to prevent diffraction from occurring in the diffraction grating 12a, which is the first infrared blocking pattern forming layer, the field of view ξ as seen from the diffraction grating 12b, which is the second infrared blocking pattern forming layer, is calculated from the aperture width w of the diffraction grating 12a, which is the first infrared blocking pattern forming layer, and the thickness d of the diffraction grating 12a, which is the first infrared blocking pattern forming layer. Angle of view ξ=2×arctan(w / 2d)

[0128] Next, we will explain the flow of the detailed image generation process in the calculation unit 26.

[0129] Here, Figure 35 is a flowchart showing the flow of the detailed image generation process. As shown in Figure 35, first, the calculation unit 26 generates and monitors a low-resolution image using the pattern example shown in Figure 5 or the wider-angle pattern example shown in Figure 9 (step S41).

[0130] Next, the calculation unit 26 detects changes in brightness (such as movement to adjacent pixels) in the low-resolution image (step S42).

[0131] Next, the calculation unit 26 uses diffraction gratings 12a and 12b to narrow down the field of view at the center of the pixel where the fluctuation was detected (step S43).

[0132] Here, Figure 36 shows another example of the infrared light receiving unit 17. As shown in Figure 36, the infrared light receiving unit 17 can change the direction of the center of the field of view by shifting the phase of the diffraction grating 12a, which is the first infrared blocking pattern forming layer.

[0133] Next, the calculation unit 26 sets a grid pattern for a narrow field of view and high resolution in the second layer (step S44).

[0134] Figure 37 shows yet another example of the infrared light receiving unit 17. In the example shown in Figure 37, there are two infrared blocking pattern forming layers. Figure 37(a) is a top view of the infrared light receiving unit 17, and Figure 37(b) is a cross-sectional view of the infrared light receiving unit 17.

[0135] As shown in Figure 37, it is also possible to set a grid pattern (non-periodic random pattern) for high resolution with a narrow field of view on the diffraction grating 12b, which is the second layer for forming the infrared blocking pattern. In the example shown in Figure 37, the size and spacing of the apertures of the diffraction gratings 12a and 12b are random and mixed. This allows for a mixture of monitoring of a specific direction with high resolution and monitoring of a wide area with low resolution. Large apertures and spacing make it easier to detect a wide area with coarse resolution. Small apertures and spacing make it easier to detect a narrow area with fine resolution.

[0136] Returning to Figure 35, the calculation unit 26 generates a high-resolution image of the target field of view (step S45).

[0137] Furthermore, it is possible to further refine the field of view and resolution by adding more layers to the diffraction grating of the infrared light receiving unit 17.

[0138] Here, Figure 38 shows the result of narrowing the dominant wavelength by switching the infrared blocking pattern of the diffraction grating over time. Figure 38 shows an example of a diffraction waveform from a specific field of view (θ=0°). In Figure 38, the diffraction grating 12b, which is the second layer forming the infrared blocking pattern, has a thickness of 50 μm and a repetition period of 100 μm. The example shown in Figure 38 is one in which the infrared blocking patterns of the two layers of diffraction gratings 12a and 12b are switched over time to narrow the dominant wavelength.

[0139] For example, the wavelength at which infrared radiation reaches its maximum intensity from an object at room temperature (310K) is approximately 10 μm. On the other hand, the wavelength at which infrared radiation reaches its maximum intensity from high temperatures such as flames (several hundred to around 1000°C) is approximately 5 μm to 2 μm. It is known that the shorter the infrared wavelength, the narrower the minimum vibration interval of the diffraction pattern tends to be, and the more concentrated the vibration spread tends to be. Furthermore, it is known that at the same wavelength, if the aperture width of the diffraction gratings 12a and 12b is larger than the wavelength, the diffraction range becomes smaller.

[0140] Therefore, the infrared blocking patterns of the two-layer diffraction gratings 12a and 12b are switched over time to narrow down the direction of high infrared intensity, and then the aperture width of the diffraction gratings 12a and 12b is narrowed to less than 10 μm, or about 1 μm, to separate the waveforms. In this way, high-temperature objects such as flames can be detected.

[0141] Thus, according to this embodiment, a high-resolution image in a specific oblique direction can be detected using the diffraction patterns of the first and second layers of the diffraction grating, which are adjusted so that the incident light beam from direction 0° does not enter the infrared image sensor 11, and the corresponding conversion map (a table of integral kernels that performs an integral transformation from the signal output (a first light intensity function on the xy coordinates of the infrared image sensor) to the infrared image (a second light intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor)).

[0142] (Third embodiment) Next, a third embodiment will be described.

[0143] The third embodiment differs from the first embodiment in that the sensor device is attached to the ceiling light cover that encloses the fluorescent lamp. In the following description of the third embodiment, the description of parts that are the same as those of the first embodiment will be omitted, and the parts that differ from the first embodiment will be described.

[0144] Figure 39 shows an example of a sensor device 10 and AC power line 31 according to the third embodiment. Figure 39(a) is a cross-sectional view of the sensor device 10 and AC power line 31, and Figure 39(b) is a view of the power supply coupler 18b on the receiving side of the sensor device 10 from the back.

[0145] The example shown in Figure 39 illustrates the attachment of a sensor device 10 to a ceiling lighting cover 81 that covers a fluorescent lamp 80. In this case, a communication line 32 and a power supply coupler 38b are provided inside the ceiling lighting cover 81. For example, the power supply coupler 38b is constructed by bonding a high-permeability sheet 73a from the top of the ceiling lighting cover 81 and a high-permeability sheet 73b from the side of the ceiling lighting cover 81 to a high-permeability block 72 on which a power supply electromagnetic coil wire 71 as a closed-curve type power supply wiring is arranged, from both the top and bottom directions. The power supply electromagnetic coil wire 71 is assumed to receive power from a power line 83 via a power distributor 82. [Explanation of symbols]

[0146] 1. Sensor System 10 Sensor device 11. Infrared image sensor 12, 12a, 12b diffraction gratings 16, 16a, 16b Spacers 18b Power supply coupler on the receiving side 19 Adhesive Sheets 20 Arithmetic unit 23a Anomaly detection unit 38b Power supply coupler on the transmission side 61. Electromagnetic coil wire on the receiving side 63 Ferromagnetic Sheet 71. Transmission side electromagnetic coil wire

Claims

1. A computing device that calculates an infrared image of the front side of an infrared image sensor using the signal output of an infrared image sensor having a diffraction grating installed on its surface, The system includes a transformation map comprising multiple integral kernel numerical tables that perform an integral transformation from a first light reception intensity function on the x-y coordinates of the infrared image sensor to a second light reception intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor. The conversion map is switched according to the application to calculate the necessary infrared image. The system includes an anomaly detection unit that detects objects posing a safety threat from the obtained infrared image, or detects that the ambient temperature within the field of view is such that monitoring is not possible. A computing device characterized by the following features.

2. The anomaly detection unit detects the presence of a moving object within the field of view from the time change of the calculated infrared image, and detects the movement of the target object using the conversion map for detecting a high-resolution image of only the angular direction in which the detected object is located. The computing device according to feature 1.

3. The anomaly detection unit monitors intensity changes using low-resolution images, and then confirms the target of the change using high-resolution images. The computing device according to feature 2.

4. A program that causes a computer to function as a computing device that calculates an infrared image of the front side of an infrared image sensor using the signal output of an infrared image sensor with a diffraction grating installed on its surface, The function calculates the necessary infrared image by switching between multiple integral kernel numerical tables, which are used as conversion maps depending on the application, to perform an integral transformation from a first light reception intensity function on the x-y coordinates of the infrared image sensor to a second light reception intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor. The system has a function to detect objects that pose a safety threat from the obtained infrared image, or to detect that the ambient temperature within the field of view is such that monitoring is not possible. A program characterized by implementing the above using the computer.

5. An anomaly detection method performed by a computing device, The calculation device takes as input a first light reception intensity function on the x-y coordinate of an infrared image sensor, which is the signal output of an infrared image sensor having a diffraction grating installed on its surface, and, assuming that the electromagnetic waves incident on the infrared image sensor are predominantly of a specific wavelength band, calculates a second light reception intensity function on the θ-φ coordinate of the direction of the incident light beam to the infrared image sensor. The calculation device performs the step of outputting whether or not there is an abnormality regarding the temperature on the front side of the infrared image sensor, An anomaly detection method characterized by including

6. A calculation method performed by a computing device that calculates an infrared image of the front side of an infrared image sensor using the signal output of an infrared image sensor having a diffraction grating installed on its surface, The calculation device includes a step of providing a transformation map which contains a plurality of integral kernel numerical tables that perform an integral transformation from a first light reception intensity function on the x-y coordinates of the infrared image sensor to a second light reception intensity function on the θ-φ coordinates of the direction of the incident light beam to the infrared image sensor, The calculation device performs the steps of switching the conversion map according to the application and calculating the necessary infrared image, The calculation device includes the steps of detecting an object that poses a safety threat from the obtained infrared image, or detecting that the ambient temperature is such that monitoring within the field of view is not possible, A calculation method characterized by including the following.

Images