Infrared image output device

Abstract

To solve the problem that screening of visitors to a building using a non-contact body surface thermometer that uses infrared rays as a countermeasure against infectious disease may invade privacy as infrared images so acquired tend to penetrate though fabric.SOLUTION: An infrared image output device is provided, configured to identify a face region and separately apply image processing rules to the face region and outside the face region to perform image processing, such that image processing can be done on human body portion other than the face and on the background and privacy can be protected not only on the human body portion but also on whereabouts.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to privacy protection technology for infrared images obtained to detect the body surface temperature of the human face or the like.

Background Art

[0002] An infectious disease caused by the novel coronavirus (COVID-19) discovered in Wuhan, China in December 2019 has spread around the world. As infection prevention measures, wearing masks and washing hands are widely recommended to the general public, and disinfectants such as medicinal ethanol and sodium hypochlorite water are installed at the entrances of hospitals, government offices, stores, etc. where people gather. In some places, non-contact body surface thermometers are also installed, and screening of people with abnormal body temperatures who are considered to have a fever is still being carried out so that infected people do not enter. In the "Law on Securing the Quality, Efficacy and Safety of Pharmaceuticals, Medical Devices, etc.", a "thermometer" is designated as a medical device, and general body surface thermometers and radiation thermometers cannot be used as thermometers. However, non-contact body surface thermometers are useful devices for simple screening of people with abnormal body temperatures as described above. A body surface thermometer senses infrared rays radiated from the exposed human body surface (such as the face) that is not covered by clothes or the like and converts them into temperature information. Since the human body surface is in contact with the outside air and is strongly affected by the outside air temperature, the body surface temperature is not the actual body temperature itself. However, there is a correlation between the body surface temperature and the actual body temperature, and it can be predicted that the higher the body surface temperature, the higher the actual body temperature, so the temperature can be compared. When comparing the surface temperatures of the exposed faces of fever patients and healthy people, it is considered that the fever patients will have a higher temperature.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When acquiring infrared images of the human body to measure its surface temperature and using them as temperature information (thermographic images), infrared radiation that penetrates clothing may be detected even in areas covered by clothing. The use of infrared images obtained from infrared radiation that has penetrated fabric could potentially lead to privacy violations. Because infrared light has a longer wavelength than visible light, it can penetrate fabrics and thin films. An 8mm video camera released by Sony Corporation in the spring of 1998 had a NIGHTSHOT function that turned off the infrared cut filter for nighttime filming of wildlife and other subjects. However, using the NIGHTSHOT function in daylight could result in infrared light from sunlight penetrating clothing and reflecting off the skin, thus capturing images. This presented a privacy issue, as the infrared light could penetrate the outer layer of clothing, revealing the outlines of fabrics that are not directly visible from the outside. Patent Document 1 discloses a technology that, as a countermeasure against the above-mentioned infringement of privacy, determines from the acquired infrared image whether a subject whose transmission should be suppressed is present, identifies the area to be blurred, and performs processing to suppress the display or recording of the infrared image in the area to be blurred, at least partially. Furthermore, it also discloses a technology that determines whether a subject whose transmission should be suppressed is present by determining whether a predetermined recognition target (a face, or a part of the human body such as a hand) is present, and performs processing to blur the area to be blurred. In Patent Document 1, the process involves two stages of processing after identifying a specific part of the human body: first, determining whether a specific part of the human body is visible in the infrared image; second, identifying the area to be blurred; and third, applying the blurring process to the identified area. This process is computationally intensive. Because the background is not processed during the blurring of the human body, the location where the infrared image was acquired can be identified by viewing the background, potentially infringing on privacy.

[0005] In this invention, when obtaining infrared images to detect the surface temperature of the human body, such as the face, the use of infrared light transmitted through clothing worn by the person to obtain the infrared image infringes on privacy. This is prevented by recognizing and identifying the face region from the infrared image, and then performing separate image processing on the recognized face region and the other two regions. The objective is to provide an infrared image output device that reduces the number of steps and performs high-speed processing for privacy protection of infrared images, including not only the human body but also the location of the person, by performing image processing after identifying the face region without identifying the other parts of the human body. [Means for solving the problem]

[0006] As the first invention, A two-dimensional infrared detection unit (A) detects infrared radiation emitted from an object, An infrared image information storage unit (B) stores infrared image information, which is an infrared image, from the infrared information detected by the 2D infrared detection unit (A), An infrared image facial region identification unit (C) recognizes the facial region of a human body from infrared image information and identifies the facial region in the infrared image information, A face region infrared image processing rule holder (D) holds face region infrared image processing rules (including a rule to not perform image processing), which are rules for image processing applied to the face region within the infrared image information. An infrared image processing rule holder (E) holds infrared image processing rules for areas other than the face region within the infrared image information (including a rule to not perform image processing), and A facial region infrared image processing unit (F) performs image processing on the facial region portion of the infrared image information based on the retained facial region infrared image processing rules, Regarding the portion of the infrared image information other than the face area, an infrared image processing unit (G) outside the face area performs image processing based on the retained infrared image processing rules outside the face area, An image processing infrared image information holding unit (H) holds the infrared image information of areas other than the face after image processing, along with the infrared image information of the face after image processing, in their original positional relationship. Image processing infrared image information output unit (J) outputs the combined image processing infrared image information, The present invention provides an infrared image output device having the following features.

[0007] As a second invention, based on the first invention, A two-dimensional temperature information holding unit (K) stores the infrared image information detected by the two-dimensional infrared detection unit (A) as two-dimensional temperature information, which is information about the two-dimensional temperature distribution. The abnormal state information output unit described later holds abnormal state judgment rules for determining an abnormal state, and An abnormal state determination unit (N) determines, using an abnormal state determination rule that is stored, whether there is a portion with a temperature above a predetermined level in the region of two-dimensional temperature information corresponding to the facial region, The present invention provides an infrared image output device according to claim 1, further comprising: an abnormal state information output unit (P) that outputs abnormal state information when the abnormal state determination unit (N) determines that an abnormal state exists; and an abnormal state information output unit (P) that outputs abnormal state information when the determination result is that an abnormal state exists. [Effects of the Invention]

[0008] The present invention enables image processing of infrared images obtained for the purpose of acquiring temperature information when screening for persons with fevers, by performing image processing on two regions: the face region and the region outside the face region. This prevents the identification of the outline of the fabric, which is not visible from the outside, when infrared rays emitted from the human body pass through the clothing being worn, thereby reducing the amount of work required and enabling high-speed image processing for privacy protection, including the location of the person. [Brief explanation of the drawing]

[0009] [Figure 1] Block diagram showing an example of the functional configuration of the infrared image output device of Embodiment 1. [Figure 2]Flowchart showing the processing flow of the infrared image output device of Embodiment 1 [Figure 3] Hardware configuration diagram of the infrared image output device of Embodiment 1 [Figure 4] Block diagram showing an example of the functional configuration of the infrared image output device of Embodiment 2 [Figure 5] Flowchart showing the processing flow of the infrared image output device of Embodiment 2 [Figure 6] Hardware configuration diagram of the infrared image output device of Embodiment 2 [Figure 7] Schematic diagram of the overall configuration of the infrared image output device of the present invention [Figure 8] Explanation diagram 1 of the image processing of the infrared image output device of the present invention [Figure 9] Explanation diagram 2 of the image processing of the infrared image output device of the present invention [Figure 10] Explanation diagram 3 of the image processing of the infrared image output device of the present invention [Figure 11] Explanation diagram 4 of the image processing of the infrared image output device of the present invention [Figure 12] Explanation diagram of an example of the gradation processing pattern of the infrared image output device of the present invention [Figure 13] Configuration example of connecting the infrared image output device of the present invention to a network

Modes for Carrying Out the Invention

[0010] Note that each part described below can be realized as an operation by a combination of hardware and software. Specifically, if a computer is used, hardware components such as a CPU, main memory, bus, or secondary storage device (non-volatile memory such as flash memory and SSD, storage media such as CD and DVD, and their reading drives), input devices used for information input, infrared cameras, infrared detection elements, display devices such as printing equipment and monitors, and other external peripheral devices, as well as interfaces for those external peripheral devices, communication interfaces, driver programs for controlling those hardware, and other application programs, application programs for user interfaces, etc. can be mentioned. And through the arithmetic processing of the CPU according to the program deployed on the main memory, data input from input devices and other interfaces, etc. and held on the memory or hard disk is processed and stored, or commands for controlling the above-mentioned various hardware and software are generated. Or the functional blocks of this device may be realized by dedicated hardware.

[0011] In addition, each embodiment described in this specification can be realized not only as an operation method but also as a device in part or in whole. Also, a part of such a device can be configured as software. Furthermore, software products used to cause a computer to execute such software, and recording media on which the products are fixed are of course included in the technical scope of each embodiment described in this specification (the same applies throughout this specification).

[0012] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Note that the present invention should not be limited to these embodiments and can be implemented in various modes without departing from the gist thereof.

[0013] <Outline of Infrared Image Output Device>

[0014] FIG. 7 is a schematic diagram of the overall configuration of the infrared image output device of the present invention. Figure 7 shows an infrared image output device that captures the area including the visitor's (0713) face within the imaging range, detects infrared light passing through an infrared lens (0714) with an infrared detector (0715), converts it into a two-dimensional image with an image processing device, performs face area recognition processing, performs predetermined image processing on the face area and the area outside the face area, outputs the thermographic image to a monitor (0718) via a coaxial cable or video cable, and outputs abnormal condition information to an alarm device (0717) if it is determined that there is a part of the face area that exceeds a predetermined threshold.

[0015] <Infrared lens> Infrared lenses use materials that transmit infrared light within the wavelength range detected by infrared detectors. Glass primarily transmits visible light and transmits up to near-infrared light, but its transmittance decreases in the mid-wavelength infrared region above 2 μm. Infrared lens materials used include zinc sulfite (ZnS), chalcogenides (amorphous materials composed of compounds of five elements: oxygen, sulfur, selenium, tellurium, and polonium), and zinc selenide (ZnSe). When detecting infrared radiation emitted from the human body, the wavelength of infrared radiation corresponding to the measurement range of approximately 0 to 50°C falls within the range of 10.4 μm to 8.9 μm, according to Wien's displacement law equation described later. Therefore, germanium, ZnS, and silicon are used. The choice of lens material can be made appropriately depending on its price and performance.

[0016] <Infrared detection element> As an infrared detection element, near-infrared regions can be detected even in video cameras and digital cameras that use CCD elements, whose primary purpose is to receive visible light, by removing the infrared cut filter. There are two types of thermal imaging detectors used to detect mid-wavelength and far-infrared radiation emitted from objects for purposes such as measuring radiant temperature: thermal detectors (uncooled) and quantum detectors (cooled). Thermal detectors measure temperature changes in a sensor caused by incoming infrared radiation by detecting changes in physical phenomena such as changes in electrical resistance, thermoelectric power, and pyroelectric effect, converting these into electrical signals for output. Thermal coupling is a phenomenon in which sensors try to match each other's temperatures, but since each sensor is very small, its own temperature has almost no effect. The other type, quantum detectors, are sensors that directly convert incoming infrared radiation into quanta. They measure by converting infrared photons into electrical signals proportional to the number of grid cells. They have the advantage of faster response and higher sensitivity than thermal detectors, but they are susceptible to thermal noise such as heat emitted by the measurement circuit itself, which can cause temperature errors, so they basically need to be cooled to a sufficiently low temperature with liquid nitrogen or similar.

[0017] For measuring radiation temperature from the surface of the human body, thermal detectors are perfectly usable due to their characteristics (response speed, etc.) and the fact that they do not require cooling, making them suitable for inexpensive radiation thermometers. Thermal detectors include bolometers, pyrometers, and thermocouples, but microbolometers, which use MEMS technology to arrange bolometers in a two-dimensional structure, are frequently used. These are manufactured by using microfabrication technology to erect columns called stats on a silicon substrate, forming thin films of heat-sensitive elements made of vanadium oxide or amorphous silicon in the air, and creating a two-dimensional grid. The infrared detection element used in the infrared image output device of the present invention is not limited to bolometers. The effects of the present invention can be obtained even if other infrared detection elements are used as appropriate within the design scope.

[0018] <Image Processing> When infrared light strikes a thin film of vanadium oxide or amorphous silicon, which is the thermal element of a bolometer, its temperature changes vertically. This temperature change alters the resistance, and by passing an electric current through it and measuring this resistance, the amount of infrared energy can be measured. From the correlation between the amount of infrared energy and temperature, the signal from the microbolometer is calculated as the amount of infrared energy, and this is then mapped to a thermal image representing the apparent temperature distribution using brightness gradation, contrast, or a color pattern.

[0019] Any object with a temperature emits electromagnetic waves from its surface. An object with 100% surface emissivity is called a perfect blackbody, and there is a relationship between the spectrum of electromagnetic waves emitted by such a perfect blackbody and its surface temperature, known as Planck's radiation law. Planck's law is a formula in physics that describes the wavelength distribution of spectral radiance, or energy density, of electromagnetic waves emitted from a blackbody. Any object emits electromagnetic waves (i.e., light) according to its temperature. When spectral radiance is graphed for each temperature based on Planck's law (horizontal axis: wavelength, vertical axis: spectral radiance), the curve is convex upwards. Wien's displacement law shows the relationship between the position of the peak for each temperature. In other words, Wien's displacement law shows the relationship between temperature and the wavelength at which spectral radiance is maximum at each temperature.

[0020] Since the human body temperature is around 36°C, according to Wien's displacement law (λ (wavelength) = 0.00289777 / T (temperature)), the infrared energy peaks at a wavelength of 9.4 μm at 36°C (309 K). Therefore, if infrared radiation around 9.4 μm can be detected, infrared radiation from the human body can be detected efficiently. However, the temperature of the human body surface rarely shows the standard body temperature of 36.5°C measured with a thermometer in areas close to the body, such as the armpit or inside the mouth. This is because it is strongly influenced by the outside air. In winter, the body surface temperature decreases due to heat dissipation from exposure to the outside air, and in summer, the body surface temperature rises more than in winter due to the influence of the outside temperature. From the correlation between body surface temperature and actual body temperature, which is separately determined from the infrared radiation energy of the body surface, a function can be created to estimate actual body temperature from the body surface temperature. In order to detect a wide area including the atmosphere around the human body and the human body itself, it is necessary to be able to detect a temperature range of about 0 to 50°C. To detect this temperature range, it is necessary to be able to detect infrared light in the wavelength range of 8.9 to 10.4 μm, according to Wien's displacement law.

[0021] <Embodiment 1> Mainly Claim 1

[0022] <Overview of Embodiment 1> A two-dimensional infrared detection unit detects infrared radiation emitted from an object, and from the resulting infrared image, it recognizes the facial region of the human body. After performing separate image processing on the facial region and the area outside the facial region, the images are combined and output.

[0023] <Embodiment 1 Functional Configuration> Figure 1 is a block diagram showing an example of the functional configuration of the infrared image output device of this embodiment. As shown in the figure, the infrared image output device (0100) of the present invention includes a two-dimensional infrared detection unit (A) (0101), an infrared image information holding unit (B) (0102), an infrared image area identification unit (C) (0103), an infrared image processing rule holding unit (D) (0104), an infrared image processing rule holding unit (E) (0105) outside the facial area, an infrared image processing processing unit (F) (0106), an infrared image processing unit (G) (0107) outside the facial area, an infrared image information holding unit after image processing (H) (0108), and an infrared image information output unit (J) (0109) after image processing.

[0024] <Embodiment 1: Two-dimensional infrared detection unit (A) (0101)> The two-dimensional infrared detection unit (A)(0101) is configured to detect infrared radiation emitted from an object.

[0025] As described above, the two-dimensional infrared detection unit (0101) consists of an infrared lens optical system and an infrared detector, etc., as an infrared camera. Relatively inexpensive devices such as body surface thermometers use the above-mentioned thermal detectors as infrared detectors, which include bolometers, pyrometers, and thermocouples. Among these, microbolometers, which use MEMS technology to arrange bolometers in two dimensions, are often used.

[0026] <Embodiment 1 Infrared Image Information Holding Unit (B) (0102)> The infrared image information holding unit (B)(0102) is configured to hold infrared image information, which is an infrared image, from the infrared information detected by the two-dimensional infrared detection unit (A)(0101). If the infrared image information you intend to retain includes Exif (Exchangeable Image File Format) information, it is preferable to retain the image without the Exif information. This is because the presence of a person in the image at the location where the infrared image was taken constitutes a type of privacy information.

[0027] <Embodiment 1: Infrared image facial area identification unit (C) (0103)> The infrared image facial region identification unit (C)(0103) is configured to recognize the face of a human body from infrared image information and to identify the facial region in the infrared image information.

[0028] Figure 8 shows an example of body surface temperature monitoring using the infrared image output device of the present invention. Three people (0820, 0821, 0822) wearing masks are walking down a corridor towards the viewer. Figure 9 is an example of an image of this scene captured by infrared light. Although all three are wearing masks, the facial region is detected using known AI-learned facial recognition technology, based on the eyes, nose, and facial contours visible above the mask. In Figure 9, the detected and identified facial region is shown as a dotted rectangle (0923, 0924). The facial region can be selected as appropriate, such as an ellipse or a region following the facial contour, rather than a rectangle as shown in Figure 9. The two people on the left side of Figure 9 (0920, 0921) are walking side by side, one behind the other, and their facial regions overlap. In cases where the faces overlap in this manner, or when the faces do not overlap but the facial area of ​​one person (the rectangular area shown in (0924)) overlaps or touches, multiple people are identified as a single area, as shown in the rectangular frame in (0923).

[0029] <Embodiment 1: Face Region Infrared Image Processing Rule Holding Unit (D) (0104)> The facial region infrared image processing rule holding unit (D)(0104) is configured to hold facial region infrared image processing rules (including rules for not performing image processing), which are rules for image processing applied to the facial region within the infrared image information.

[0030] The system maintains rules for what kind of image processing to perform on the facial region in the acquired infrared image. For example, an infrared image acquired by the infrared detection unit (A) can be maintained in a state represented by brightness gradations corresponding to the intensity of infrared energy. Alternatively, it can be maintained as an apparent temperature distribution based on a predetermined correlation between infrared energy and temperature. As an example of image processing rules for a specific facial region within the maintained infrared image information, we will explain using an example of a black and white image processing rule. For example, there are rules to change the setting of the number of brightness gradations for a range of infrared energy, rules to change to a non-uniform gradation setting that finely divides only a specific range of infrared energy, and rules to leave the default settings unchanged. For example, if infrared image information is maintained as a temperature distribution, the range from 0°C to 50°C may have a brightness gradation setting where 50°C is 100% white and it gets darker by 1 gradation for every 1°C decrease in temperature, but there is a rule to change the brightness gradation setting to one gradation for every 0.1°C. An example of a non-uniform approach is to divide the range from 0°C to 50°C into 79 gradations, as shown in Figure 12, where the pitch increases by one degree for every 10°C rise from 0°C to 20°C, changes by one degree for every 5°C rise from 20°C to 30°C, changes by one degree for every 1°C rise from 30°C to 34°C, increases by one degree for every 0.1°C rise from 34°C to 40°C, and increases by one degree for every 1°C rise from 40°C to 50°C.

[0031] Graphics processing, such as brightness gradation, uses elements like graphics drivers. Therefore, it is preferable for the number of brightness gradations to be a power of 2, such as 64 or 128. As mentioned above, for the 79 gradations, the brightness gradation settings, which divide the image from white to black into 128 gradations, can be used discontinuously, such as skipping every other gradation, rather than continuously. The assignment of brightness gradation settings can be configured as appropriate.

[0032] The above example uses an image processing rule that changes the brightness gradation settings assuming a black and white screen, but image processing can also be performed as a color pattern, as in a typical thermographic image. High temperatures are assigned to red and low temperatures to blue, and the sequence of visible light colors from red to blue is assigned to the temperature distribution. If the initial color pattern is blue for 0°C and red for 50°C, it can be used as an infrared image processing rule for the face area that changes the color assignment to temperature. Alternatively, to emphasize high-temperature areas, a rule combining brightness gradation and color patterns can be used, such as representing the temperature of areas below a predetermined temperature with brightness gradation in black and white, and only coloring areas above the predetermined temperature red to represent them as a color pattern. An example of a predetermined temperature is 37°C, which can be used to screen for people with fevers. This should be adjusted according to the ambient temperature and the specifications and performance of the infrared detector. It is also possible to have a rule that does not perform image processing on the face area, leaving it as it was when acquired by the 2D infrared detection unit and stored in the infrared image information storage unit.

[0033] Furthermore, it is preferable that the infrared image processing rules for the face area include rules for adding a border to indicate the contour of the face area, and for changing the brightness gradation, color, line type (such as dotted or solid line), and line width of the border according to the highest temperature within the face area. For example, the brightness gradation (or color) of the border can be set according to the same rules as the highest temperature within the face area. Then, if the image includes an area with a temperature above a predetermined level (such as 37°C, assuming a person with a fever), the rule can be used to change the border to a dotted line if the temperature is below the predetermined level, and to make it a thicker solid line if the temperature is above the predetermined level.

[0034] When processing images, it is also possible to configure the system to dynamically fine-tune the image processing rules by using a separate thermometer to measure the external temperature, and using the external temperature measured by the thermometer as reference temperature information, thereby reducing the number of tone divisions, such as assigning only one tone to areas with temperatures lower than the reference temperature information. Alternatively, the reference temperature information may be the external temperature value plus or minus a predetermined value. Instead of using a separate thermometer, the average temperature or minimum temperature of an area on the screen where the temperature does not change much during the acquisition of infrared image information for about 10 seconds may be used as the reference temperature information to perform the above processing. The reference temperature information may be acquired as part of the initial startup operation when the infrared image output device of the present invention is powered on, or a separate clock unit may be provided to automatically update the reference temperature information periodically.

[0035] <Embodiment 1: Infrared image processing rule holding unit outside the facial area (E) (0105)> The infrared image processing rule holding unit (E)(0105) outside the face region is configured to hold infrared image processing rules outside the face region (including a rule that no image processing should be performed), which are rules for image processing applied to areas other than the face region in the infrared image information.

[0036] This is an image processing rule for infrared image information outside the face region identified by the infrared image face region identification unit (C). It is the area outside the dotted rectangle (0923, 0924) surrounding the face region of the three people shown in Figure 9, and includes all parts of the three people's clothed bodies, exposed hands, and the corridor wall. As an example of an infrared image processing rule outside the face region, the rule for image processing of a black and white infrared image will be explained as an example. Figure 12 shows two examples of brightness gradation processing patterns outside the face region. Pattern 1 assumes that the surface of clothing is slightly warmed by the infrared radiation emitted by the human body, so in cool periods with an outside temperature of around 15°C, two types of brightness gradations (or two types of colors) are assigned with 20°C as the boundary. Pattern 2 represents the entire measurement range (0°C to 50°C in the example of Figure 12) with one gradation (or one color). In the example in Figure 9, the outlines of clothing collars and hands are also included for illustrative purposes, but in Pattern 1, as shown in Figure 10, the image is divided into two parts: an area that is almost the outline of the human body and the background of the building. In the case of Pattern 2, as shown in Figure 11, the entire area outside the face is filled with one tone (or one color). Although not shown in Figure 12 as an example of a pattern, it is also possible to set a rule for image processing such that, considering the influence of outside temperature, the temperature range that can be considered from the range of infrared energy emitted by a clothed human body (for example, from the equivalent of outside temperature to about 40°C, or above the equivalent of outside temperature) is set to one tone or coarse tone (one color or reduction of the color pattern), and other temperature ranges (below the outside temperature corresponding to the background area) are processed without changing the tone setting (color pattern).

[0037] Furthermore, in winter, when many people are wearing thick clothing and the outside temperature is low, the likelihood of privacy violations from infrared images of the human body is considered low. Therefore, a rule that does not process the image outside the face area, as it was stored in the infrared image information storage unit (B), is also acceptable. As mentioned in the facial area infrared image processing rule, it is preferable to have a rule that involves conditional judgments linked to the outside temperature, such as not processing the infrared image outside the face area when the outside temperature or the temperature of the unchanging background area is below a predetermined temperature (e.g., a temperature equivalent to winter, 5°C or below). However, location information, such as where a person is in the acquired infrared image, is a type of privacy. From the perspective of privacy protection, it is desirable to process the image outside the face area as well, so that the location cannot be identified.

[0038] It is preferable that both the infrared image processing rules for the facial region and the infrared image processing rules for areas outside the facial region be configured to be readable and addable from outside the infrared image output device via a USB terminal, LAN terminal, RS-485 terminal, etc. It is also preferable that the image processing rules can be deleted and written (saved) on the infrared image output device. It is even more preferable that the contents of the image processing rules be viewable.

[0039] Only the facial region is identified in the image, and the area outside the facial region is processed based on infrared image processing rules for areas outside the facial region. This requires less processing effort than Patent Document 1, which identifies the body part after recognizing it and then blurs it. While rules for uniformly filling in or mosaic processing the entire area outside the facial region are also possible, a key feature is that the system can be configured to perform processing according to the infrared energy (or temperature distribution obtained from correlation) of each point, as described above.

[0040] <Embodiment 1: Facial Region Infrared Image Processing Unit (F) (0106)> The facial region infrared image processing unit (F)(0106) is configured to perform image processing on the portion of the facial region within the infrared image information based on the retained facial region infrared image processing rules.

[0041] When performing image processing on the facial region, the specified facial region may be extracted from the original infrared image information, and image processing may be performed on the extracted infrared image information of the facial region. Alternatively, the facial region may not be extracted from the original image, and image processing may be performed only within the specified facial region. Image processing is performed according to the selected image processing rule from the above-mentioned infrared image processing rules for the face area. The infrared image processing rules for the face area are stored in the infrared image output device by methods such as storing the settings in non-volatile memory, and a pre-set infrared image processing rule for the face area is selected at startup, or selected manually from the stored infrared image processing rules for the face area while the device is running. When the infrared image processing rule for the face area is changed during operation, the newly selected infrared image processing rule for the face area may be used until the power is turned off, and the setting may be erased and reverted to the previous setting when the power is turned off, or the setting stored in non-volatile memory may be overwritten when a new setting is selected. The infrared image output device may be equipped with a monitor or similar device so that the infrared image processing rules for the face area and the infrared image processing rules for the outside of the image area described later can be edited or newly created while displaying the contents. The device may also be configured to automatically select the infrared image processing rule for the face area and perform image processing according to conditions such as the external temperature.

[0042] <Embodiment 1: Infrared image processing unit outside the facial area (G) (0107)> The infrared image processing unit (G)(0107) outside the facial region is configured to perform image processing on parts of the infrared image information other than the facial region based on the stored infrared image processing rules outside the facial region.

[0043] When performing image processing on areas outside the face region, the identified face region may be extracted from the original infrared image information, and image processing may be performed on the remaining infrared image information outside the face region. Alternatively, image processing may be performed only on the identified areas outside the face region without extracting the face region. Processing is performed according to the selected infrared image processing rule for the area outside the face, as described above. The infrared image processing rules for the area outside the face are stored in the infrared image output device by methods such as storing the settings in non-volatile memory, similar to the infrared image processing rules for the face area, and a pre-set infrared image processing rule for the area outside the face is selected at startup [Toru Watanabe 1]. Alternatively, the infrared image processing rules for the area outside the face are selected and set by methods such as manually selecting from the stored infrared image processing rules for the area outside the face while the device is running. When the infrared image processing rules for the area outside the face are changed during operation, the newly selected infrared image processing rules for the area outside the face may be used until the power is turned off, and the setting may be erased when the power is turned off, returning to the previous setting. Alternatively, when a new setting is selected, the setting stored in non-volatile memory may be overwritten. The system may also be configured to automatically select an infrared image processing rule for the area outside the face and perform image processing according to conditions such as the external temperature.

[0044] <Embodiment 1 Image Processing Infrared Image Information Holding Unit (H) (0108)> The post-processing infrared image information holding unit (H)(0108) is configured to hold the post-processing infrared image information of areas other than the face and the post-processing infrared image information of the face in their original positional relationship.

[0045] The infrared image information of the face region and the area outside the face region, after being processed according to their respective image processing rules, is stored together. This "together" does not mean that the two images (infrared image of the face region and infrared image of the area outside the face region) are stored as separate images linked together. "Together" means that if the infrared image information of the face region is extracted and image processed, the processed infrared image information of the face region is fitted into a predetermined position in the infrared image information of the area outside the face region (the original positional relationship before image processing), and stored as a single infrared image. Alternatively, this "together" means that if the infrared image information of the face region is not extracted, and image processing is performed on the face region and the area outside the face region based on the coordinates in the infrared image information while maintaining the original positional relationship, the processed infrared image information is stored as if the two image processing operations for the face region and the area outside the face region were performed together as a single infrared image.

[0046] <Embodiment 1 Image Processing Infrared Image Information Output Unit (J) (0109)> The post-processing infrared image information output unit (J)(0109) is configured to output the combined post-processing infrared image information.

[0047] The infrared image information after image processing is displayed on a monitor (Figure 7, (0718)). Since there is a possibility of reviewing the infrared image information later, it is advisable to provide an infrared image information recording and storage unit to record and save it. Depending on the image resolution and recording time, which affect the image information file size, a large-capacity storage medium such as a hard disk or SSD is preferable. If a short recording time is acceptable, a memory disk or USB memory may be used. It is preferable to configure the infrared image information before image processing, which is stored in the infrared image information storage unit (B), to be deleted from the infrared image information storage unit (B) at a predetermined timing, such as after a certain period of time (a certain amount of data) has elapsed, after image processing is completed, or when the infrared image output device is powered off. This is because the infrared image information in the infrared image information storage unit (B) is infrared image information before image processing for privacy protection. If the infrared image information before image processing is to be retained, care should be taken in handling it, and it is preferable to limit the number of people who can handle it and manage it so that it cannot be easily viewed by third parties.

[0048] <Embodiment 1: Infrared Image Output Device (0100)> The infrared image output device (0100) identifies the facial region of the human body in the acquired infrared image information, maintains separate image processing rules for the facial region and the area outside the facial region for image processing of the facial region and the area outside the facial region (they may use the same image processing rules for both, or they may use rules that do not apply image processing), and is configured to maintain and output the infrared image information of the processed facial region and the area outside the facial region together.

[0049] The infrared image output device of the present invention, installed at the entrance of a building or room, is used to screen for individuals suspected of having a fever among those entering the building. A supervisor can visually check the infrared image information output to a monitor and speak to anyone who shows abnormalities. For individuals suspected of having a fever who show abnormalities, measures such as having their temperature taken separately with a thermometer capable of measuring actual body temperature or being denied entry can be taken.

[0050] <Embodiment 1: Processing Flow> Figure 2 shows the processing flow of the infrared image output device of Embodiment 1. The processing flow is as follows: The process includes a two-dimensional infrared detection step (a) (S0201), an infrared image information retention step (b) (S0202), a facial area identification step within the infrared image (c) (S0203), a facial area infrared image processing rule retention step (d) (S0204), an infrared image processing rule retention step outside the facial area (e) (S0205), a facial area infrared image processing step (f) (S0206), an infrared image processing step outside the facial area (g) (S0207), a post-processing infrared image information retention step (h) (S0208), and a post-processing infrared image information output step (j) (S0209).

[0051] The infrared image output device is The two-dimensional infrared detection step (a) (S0201) performs a process to detect infrared radiation emitted from an object. The infrared image information retention step (b) (S0202) performs a process to retain the infrared information detected in the two-dimensional infrared detection step (a) (S0201) as infrared image information, which is an infrared image. Step (c)(S0203) for identifying the facial region in the infrared image involves recognizing the facial region of the human body from the infrared image information and performing a process to identify the facial region in the infrared image information. The facial region infrared image processing rule retention step (d) (S0204) is a process that retains the facial region infrared image processing rules (including the rule of not performing image processing), which are the rules for image processing applied to the facial region in the infrared image information. Step (e) (S0205) of retaining the infrared image processing rules outside the face region is a process of retaining the infrared image processing rules outside the face region (including the rule of not performing image processing), which are the rules for image processing applied to areas other than the face region in the infrared image information. The facial region infrared image processing step (f) (S0206) performs image processing on the portion of the facial region in the infrared image information based on the retained facial region infrared image processing rules. The infrared image processing step (g) (S0207) outside the facial region performs image processing on the portion of the infrared image information other than the facial region, based on the retained infrared image processing rules outside the facial region. The post-processing infrared image information retention step (h) (S0208) performs a retention process on the infrared image information of areas other than the face after image processing, along with the infrared image information of the face after image processing. The post-image processing infrared image information output step (j) (S0209) outputs abnormal state information based on the stored abnormal state judgment rule if there is a portion of the 2D temperature information area corresponding to the face area that exceeds a predetermined temperature information. This is a method of instructing an infrared image output device to perform a series of processes like the ones described above.

[0052] <Embodiment 1 Hardware>

[0053] Figure 3 is a conceptual diagram showing an example of the hardware configuration of the infrared image output device of this embodiment. As shown in the figure, it is thought that the configuration is often similar to that of a PC, and is therefore similar to that of a PC. It consists of a "CPU", a "chipset" consisting of a northbridge and a southbridge, "non-volatile memory", "main memory", "I / O controller", "USB, IEEE1394, LAN terminal, etc.", "BIOS", "PCI slot", and "real-time clock".

[0054] The various programs and data (information) stored in non-volatile memory are loaded into main memory upon system startup, and the CPU is configured to sequentially perform calculations using the data by accepting execution instructions.

[0055] In other words, as shown in Figure 3, the infrared output device has a non-volatile memory containing, in addition to the OS (operating system) and device driver, a 2D infrared detection program, an infrared image information retention program, a facial area identification program within the infrared image, a facial area infrared image processing rule retention program, an area outside the facial area infrared image processing rule retention program, a facial area infrared image processing program, a post-processed infrared image information retention program, and a post-processed infrared image information output program. The data stored there includes infrared image information, facial area infrared image processing rules, and area outside the facial area infrared image processing rules. When this system is started on a computer, these are loaded into main memory, and upon receiving a startup command, the CPU sequentially performs calculations using the programs and data. <Effect of Embodiment 1>

[0056] According to the infrared image output device of this embodiment, a human face is recognized in the infrared image information, the face region is identified, and image processing is performed on the face region and the area outside the face region. Since the region identification for image processing is performed only once, when the face region is identified, the amount of work required can be reduced and the process can be sped up. Furthermore, since image processing is performed not only on the human body but also on the background, information about the location where the infrared image was acquired (where the person is) can also be kept confidential. Privacy protection processing can be performed simultaneously on both the human body and the location.

[0057] <Embodiment 2 Overview> Mainly Claim 2 From the acquired infrared image, the signal from the microbolometer is calculated as infrared energy based on the correlation between infrared energy and temperature. This signal is then stored as two-dimensional temperature information, mapped to a contrast or color pattern (color gradation) as an apparent temperature distribution. If a portion of the two-dimensional temperature information corresponding to the facial area exceeds a predetermined temperature, abnormal condition information is output.

[0058] <Embodiment 2 Functional Configuration> Figure 4 is a block diagram showing an example of the functional configuration of an infrared image output device of Embodiment 2 based on Embodiment 1. As shown in the figure, the infrared image output device (0400) of the present invention comprises a two-dimensional infrared detection unit (A) (0401), an infrared image information holding unit (B) (0402), an infrared image area identification unit (C) (0403), an infrared image processing rule holding unit (D) (0404), an infrared image processing rule holding unit (E) (0405) outside the facial area, and an infrared image processing unit (F) (0406) It has a facial region-external infrared image processing unit (G) (0407), an image-processed infrared image information holding unit (H) (0408), an image-processed infrared image information output unit (J) (0409), and further includes a two-dimensional temperature information holding unit (K) (0410), an abnormal state judgment rule holding unit (M) (0411), an abnormal state judgment unit (N) (0412), and an abnormal state information output unit (P) (0413).

[0059] <Embodiment 2: Two-dimensional temperature information holding unit (K) (0410)> The two-dimensional temperature information holding unit (K) (0410) is configured to hold the infrared image information detected by the two-dimensional infrared detection unit (A) (0401) as two-dimensional temperature information, which is information about the two-dimensional temperature distribution.

[0060] The infrared energy corresponding to each point in the infrared image detected by the 2D infrared detection unit is converted into temperature distribution information correlated with the infrared energy and stored as 2D temperature information applied to brightness gradation, contrast, or a color pattern. An example of a color pattern is an arrangement that follows the color order of visible light, with the lowest temperature in the measurable detection range being blue and the highest temperature being red. Alternatively, the image can be a monochrome display as described in Embodiment 1, but with the gradation and contrast adjusted. The infrared image information used to identify the facial region and to perform image processing on both the facial region and the area outside the facial region may be two-dimensional temperature information, which is information on temperature distribution.

[0061] <Embodiment 2: Abnormal State Judgment Rule Holding Unit (M) (0411)> The abnormal state determination rule holding unit (M)(0411) is configured to hold abnormal state determination rules for determining an abnormal state in the abnormal state information output unit described later.

[0062] For example, a rule might be used to determine an abnormal condition if there is a portion of the facial area in the stored 2D temperature information that is above a predetermined temperature (for the purpose of screening for people with fever). Other rules might include determining that there is a problem with the installation of the infrared image output device if there is a portion outside the facial area that is above a predetermined temperature (high heat sources such as stoves or lighting fixtures are present within the infrared image field of view), or determining that the device is malfunctioning if all areas show a temperature above a predetermined temperature.

[0063] <Embodiment 2: Abnormal State Determination Unit (N)(0412)> The abnormal state determination unit (N)(0412) is configured to determine, using stored abnormal state determination rules, whether there is a portion of the 2D temperature information region corresponding to the face region that is above a predetermined temperature.

[0064] An example of a rule for determining an abnormal condition is to screen for people suspected of having a fever. This rule uses 37°C as a threshold, and if a portion of the face area is 37°C or higher, it is considered an abnormal condition. The threshold can be adjusted based on the ambient temperature. Furthermore, it is preferable to provide information indicating an abnormal measurement environment or device malfunction, rather than just screening for people suspected of having a fever, when high heat sources such as stoves or fluorescent lights are present within the field of view of the 2D infrared detection unit (A), or when there is a concern about device malfunction. This is because if a high heat source is present within the field of view and happens to be captured in the facial area, there is a possibility of misidentification.

[0065] <Embodiment 2 Abnormal State Information Output Unit (P) (0413)> The abnormal state information output unit (P) (0413) is configured to output abnormal state information when the abnormal state determination unit (N) (0412) determines that an abnormal state exists.

[0066] An example of abnormal state information output will be explained using Figure 10. In the example in Figure 10, the predetermined temperature that is judged as an abnormal state is 37°C. Figure 10 is a display of 2D temperature information, which is 2D temperature information after image processing has been performed on the face area and the area outside the face area. The face area (1023) contains two people whose faces are overlapping, positioned one in front of the other. Because the faces overlap, it is considered a single face area (1023). In the example in Figure 10, the highest temperature within the face area (1023), 35.4°C, is displayed near the face area (1023). Because it is below 37°C, which is judged as an abnormal state, the dotted rectangle surrounding the face area is in a dark tone (in the color pattern, colors such as yellow and orange indicate temperatures below red). The highest temperature in the right-hand facial region (1024) is 38.1°C, exceeding the predetermined temperature of 37°C. Therefore, the rectangle surrounding the facial region (1024) is a solid line and displayed in a near-white gradation (a reddish color in the color pattern indicating high temperature). "NG" is displayed near the facial region (1024) (1028). Within the facial region (1024), the forehead area shows a near-white gradation indicating high temperature. When a single facial region is used, as with the two people on the left, if at least one person shows a temperature of 37°C or higher, both are treated as an abnormal state and abnormal state information is output. This is to screen for suspicious individuals. A detailed investigation (actual body temperature measurement using a thermometer) may be conducted on both individuals, or they may be asked to stand in front of the infrared image output device again with a slight distance between them to reconfirm their body surface temperature. In the latter case, the number of people who need to be asked to undergo a detailed investigation can be narrowed down.

[0067] In addition to displaying abnormal status information on a monitor, the system can also be configured to emit an alarm sound from an alarm lamp or speaker, as shown in the infrared image output device alarm (0717) in Figure 7, to draw attention. In locations where emitting an alarm sound would be problematic, the monitor display alone is likely to be overlooked, so it is advisable to at least illuminate the alarm lamp so that it is visible to the observer.

[0068] Furthermore, it is preferable to issue abnormal status information different from that for detecting a person with a fever, not only for screening people suspected of having a fever, but also when high heat sources such as heaters or fluorescent lights are present within the field of view of the 2D infrared detection unit (A), or when there is a concern about equipment malfunction. This is because if a high heat source is present, there is a possibility of misidentification if it is captured in the facial area.

[0069] <If you have an internet connection> A configuration in which multiple infrared image output devices of the present invention are connected via an internet line, and a server that records and stores the processed infrared image information is also connected to the internet, will be explained with reference to Figure 13. Figure 13 shows infrared image output devices (1330A) to (1330K) connected to the internet (1329), and a server (1331) also connected. The number of infrared image output devices and servers is not limited to those shown in Figure 13. There may be multiple servers, and the number of infrared image output devices may be 10 or more, or 10 or less, as shown. Furthermore, for illustrative purposes, the infrared image output device is shown as a camera unit (1330A1) and speaker unit (1330A2) mounted on a tripod, but the shape is not limited as long as it functions according to the configuration of the infrared image output device of the present invention.

[0070] The infrared image output device of the present invention can be installed in multiple locations via the internet, such as major train station ticket gates and major road intersections throughout Japan. By using the internet, monitors can monitor from locations far away from the infrared image output device, and multiple locations can be monitored simultaneously by arranging multiple monitors. The infrared image output device can warn individuals suspected of having a fever by outputting a voice message from a speaker attached to each device, such as, "There is a concern about abnormal body temperature; please undergo further examination (measure actual body temperature with a thermometer, perform tests such as PCR testing, visit a hospital, etc.)." Alternatively, the device may be configured to automatically output such a message to people whose faces are visible in the image when abnormal status information is output. Alternatively, the device may be configured to compare the information with facial image data on a My Number Card and notify a designated contact (email address, etc.).

[0071] Identification information (such as equipment number) for each individual infrared image output device (1330A) to (1330K) is stored in association with information about the installation location. Statistics can be obtained for the number of people detected within the field of view over a predetermined period (e.g., every day or every hour) (if multiple faces overlap and it is difficult to separate and count them, the number of facial areas may be used), and the number of people for whom abnormal status information indicating a possible fever was output (for the reasons mentioned above, the number of facial areas may also be used). Based on the acquired data, for example, to understand the COVID-19 infection situation, the effective reproduction number for each region can be calculated using a simplified calculation method discovered by a group led by Professor Takeo Kotani of the Faculty of Engineering, Tottori University, as announced in a press release on November 25, 2020 (dividing the number of confirmed cases 5 days after a certain day by the average number of people with fever over the 3 days before and after a certain day: i.e., 1 week). Based on the obtained simplified effective reproduction number, it is possible to understand the differences between regions and the magnitude of infection risk in each region. By color-coding areas based on the rate of detected fevers, a fever map can be created that visually represents areas with a high number of fevers. This allows for understanding the infection risk in each region. It can be used not only for COVID-19 but also for identifying other fever-related infectious diseases such as influenza.

[0072] After detecting a person who may have a fever (hereinafter referred to as a "fever-affected person"), if they move out of the field of view of an infrared image output device and then are captured and recognized by another infrared image output device in a different location, the system can be configured to track their movement by inheriting information detected at previously passed locations. The server (1331) stores the facial recognition information of the person detected as a fever-affected person, and tracking can be performed by comparing it with the facial recognition information recognized in various locations. Since communication congestion occurs if there is only one server over a wide area (such as Japan), it is advisable to divide the servers into one per prefecture or region, each responsible for a specific area of ​​travel such as local trains or cars. When matching facial recognition information, it is advisable to match it with the server responsible for that location as well as with servers responsible for adjacent areas. Furthermore, it is preferable to limit the scope of access permissions granted to the infrared image after image processing (only the facial area is discernible), so that only the administrator of the infrared image output device or the server administrator can view and access it. Alternatively, access should be limited to operators approved and authorized by the administrator.

[0073] <Embodiment 2: Processing Flow> Figure 5 shows the processing flow of the infrared image output device of Embodiment 2, which is based on Embodiment 1. The processing flow is as follows: The process includes a two-dimensional infrared detection step (a) (S0501), an infrared image information retention step (b) (S0502), a two-dimensional temperature information retention step (k) (S0503), a facial area identification step within the infrared image (c) (S0504), a facial area infrared image processing rule retention step (d) (S0505), an infrared image processing rule retention step outside the facial area (e) (S0506), a facial area infrared image processing step (f) (S0507), an infrared image processing step outside the facial area (g) (S0508), a post-processing infrared image information retention step (h) (S0509), a post-processing infrared image information output step (j) (S0510), an abnormal state judgment rule retention step (m) (S0511), and an abnormal state information output step (p) (S0512).

[0074] The infrared image output device is The two-dimensional infrared detection step (a) (S0501) performs a process to detect infrared radiation emitted from an object. The infrared image information retention step (b) (S0502) performs a process to retain the infrared information detected in the two-dimensional infrared detection step (a) (S0501) as infrared image information, which is an infrared image. The 2D temperature information retention step (k) (S0503) performs a process to retain the infrared image information detected in the 2D infrared detection step (a) (S0501) as 2D temperature information, which is information about the 2D temperature distribution. Step (c)(S0503) for identifying the facial region in the infrared image involves recognizing the facial region of the human body from the infrared image information and performing a process to identify the facial region in the infrared image information. The facial region infrared image processing rule retention step (d) (S0504) is a process that retains the facial region infrared image processing rules (including the rule of not performing image processing), which are the rules for image processing applied to the facial region in the infrared image information. Step (e) (S0505) of retaining the infrared image processing rules outside the face region is a process of retaining the infrared image processing rules outside the face region (including the rule of not performing image processing), which are the rules for image processing applied to areas other than the face region in the infrared image information. The facial region infrared image processing step (f) (S0506) performs image processing on the portion of the facial region in the infrared image information based on the retained facial region infrared image processing rules. The infrared image processing step (g) (S0507) outside the facial region performs image processing on the portion of the infrared image information other than the facial region, based on the retained infrared image processing rules outside the facial region. The post-processing infrared image information retention step (h) (S0508) involves retaining the infrared image information of areas other than the face after image processing, along with the infrared image information of the face after image processing. The post-image processing infrared image information output step (j) (S0509) performs a process to output abnormal state information based on the stored abnormal state judgment rule if there is a portion of the 2D temperature information area corresponding to the face area that exceeds a predetermined temperature information. The abnormal state determination rule retention step (m) (S0511) performs the process of retaining the abnormal state determination rule for determining an abnormal state in the abnormal state information output step (p) (S0512) described below. The abnormal state information output step (p) (S0512) outputs abnormal state information based on the stored abnormal state judgment rule when a portion of the 2D temperature information area corresponding to the face area exceeds a predetermined temperature information. This is a method of instructing an infrared image output device to perform a series of processes like the ones described above.

[0075] <Embodiment 2 Hardware>

[0076] Figure 6 is a conceptual diagram showing an example of the hardware configuration of an infrared image output device of Embodiment 2, which is based on Embodiment 1. As shown in the figure, it is thought that the configuration is often similar to that of a PC, and is therefore similar to that of a PC. It consists of a "CPU", a "chipset" consisting of a northbridge and a southbridge, "non-volatile memory", "main memory", "I / O controller", "USB, IEEE1394, LAN terminal, etc.", "BIOS", "PCI slot", and "real-time clock".

[0077] The various programs and data (information) stored in non-volatile memory are loaded into main memory upon system startup, and the CPU is configured to sequentially perform calculations using the data by accepting execution instructions.

[0078] In other words, as shown in Figure 6, the infrared output device has a non-volatile memory containing, in addition to the OS (operating system) and device drivers, a 2D infrared detection program, an infrared image information retention program, a facial area identification program within the infrared image, a facial area infrared image processing rule retention program, an area outside the facial area infrared image processing rule retention program, a facial area infrared image processing program, a post-processed infrared image information retention program, a post-processed infrared image information output program, a 2D temperature information retention program, an abnormal state judgment rule retention program, and an abnormal state information output program. The data stored includes infrared image information, facial area infrared image processing rules, area outside the facial area infrared image processing rules, 2D temperature information, abnormal state judgment rules, and abnormal state information. When this system is started on a computer, these are loaded into main memory, and upon receiving a startup command, the CPU sequentially performs calculations using the programs and data. <Effects of Embodiment 2>

[0079] According to the infrared image output device of this second embodiment, if a person is captured in the image showing an abnormal condition with a temperature exceeding a predetermined temperature in the two-dimensional temperature information of the face area, an abnormality can be detected. Based on Embodiment 1, image processing for privacy protection is performed at high speed, and since the face area can be identified, the target person can be quickly found among people standing in front of the two-dimensional infrared detection unit of the infrared image output device. The background area can also be image processed simultaneously with the rest of the human body, and privacy protection processing can be performed on that area as well. [Explanation of symbols]

[0080] 0100 Infrared Image Output Device 0101 2D infrared detection unit (A) 0102 Infrared image information storage unit (B) 0103 Infrared image facial area identification unit (C) 0104 Facial area infrared image processing rule retention unit (D) 0105 Infrared image processing rule retention unit outside the facial area (E) 0106 Facial Region Infrared Image Processing Unit (F) 0107 Extra-facial infrared imaging processing unit (G) 0108 Image processing infrared image information storage unit (H) 0109 Image processing infrared image information output unit (J)

Claims

1. a two-dimensional infrared detection unit (A) that detects infrared rays emitted from an object; an infrared image information storage unit (B) that stores infrared image information, which is an infrared image, based on the information of the infrared rays detected by the two-dimensional infrared detection unit (A); an infrared image face region identifying unit (C) that recognizes a face of a human body from infrared image information and identifies a face region in the infrared image information; a face region infrared image processing rule storage unit (D) for storing face region infrared image processing rules (including a rule that image processing is not performed) that are rules for image processing to be performed on face regions in infrared image information; a non-face area infrared image processing rule storage unit (E) that stores non-face area infrared image processing rules (including rules that do not perform image processing) that are rules for image processing to be applied to areas other than the face area in the infrared image information; a face area infrared image processing unit (F) for processing a face area in the infrared image information based on a face area infrared image processing rule stored therein; a non-face area infrared image processing unit (G) for performing image processing on a portion other than the face area in the infrared image information based on the stored non-face area infrared image processing rule; a post-image-processing infrared image information storage unit (H) for storing the post-image-processing infrared image information of the non-face region and the post-image-processing infrared image information of the face region in their original positional relationship; an image-processed infrared image information output unit (J) that outputs the combined image-processed infrared image information; An infrared image output device having:

2. a two-dimensional temperature information storage unit (K) that stores the infrared image information detected by the two-dimensional infrared detection unit (A) as two-dimensional temperature information, which is information on two-dimensional temperature distribution; an abnormal state determination rule storage unit (M) that stores an abnormal state determination rule for determining an abnormal state in an abnormal state information output unit (described later); an abnormal condition determination unit (N) that determines whether or not there is a part having a temperature equal to or higher than a predetermined temperature in the area of ​​the two-dimensional temperature information corresponding to the face area, using stored abnormal condition determination rules; 2. The infrared image output device according to claim 1, further comprising an abnormal state information output unit (P) for outputting abnormal state information when the judgment result of the abnormal state judgment unit (N) is that an abnormal state exists.

3. a two-dimensional infrared detection step (a) of detecting infrared rays emitted from an object; an infrared image information storage step (b) for storing the infrared information detected in the two-dimensional infrared detection step (a) as infrared image information, which is an infrared image; a step (c) of identifying a face region in an infrared image by recognizing a face region of a human body from the infrared image information and identifying a face region in the infrared image information; a face region infrared image processing rule storing step (d) for storing face region infrared image processing rules (including a rule that no image processing is performed) which are rules for image processing to be performed on face regions in the infrared image information; a step (e) of storing infrared image processing rules for outside the face area, which are rules for image processing to be applied to areas other than the face area in the infrared image information (including a rule that no image processing is performed); a face region infrared image processing step (f) for processing a face region in the infrared image information based on the stored face region infrared image processing rule; a non-face area infrared image processing step (g) for processing an area other than the face area in the infrared image information based on the stored non-face area infrared image processing rule; a post-image-processing infrared image information storing step (h) for storing the post-image-processing infrared image information of the non-face region and the post-image-processing infrared image information of the face region in their original positional relationship; an image-processed infrared image information output step (j) for outputting the integrated image-processed infrared image information; A method for operating an infrared image output device that is a computer having the above-mentioned steps.

4. a two-dimensional temperature information storage step (k) of storing the infrared image information detected in the two-dimensional infrared detection step (a) as two-dimensional temperature information, which is information on two-dimensional temperature distribution; an abnormal state determination rule storage step (m) for storing an abnormal state determination rule for determining an abnormal state in the abnormal state information output step (p) described later; an abnormal condition determination step (n) of determining whether or not there is a part with a temperature equal to or higher than a predetermined temperature in the area of ​​the two-dimensional temperature information corresponding to the face area, using a stored abnormal condition determination rule; 4. The method for operating an infrared image output device that is a computer according to claim 3, further comprising: an abnormal state information output step (p) of outputting abnormal state information when the judgment result in the abnormal state judgment step (n) is that an abnormal state exists.

5. a two-dimensional infrared detection step (a) of detecting infrared rays emitted from an object; an infrared image information storage step (b) for storing the infrared information detected in the two-dimensional infrared detection step (a) as infrared image information, which is an infrared image; a step (c) of identifying a face region in an infrared image by recognizing a face region of a human body from the infrared image information and identifying a face region in the infrared image information; a face region infrared image processing rule storing step (d) for storing face region infrared image processing rules (including a rule that no image processing is performed) which are rules for image processing to be performed on face regions in the infrared image information; a step (e) of storing infrared image processing rules for outside the face area, which are rules for image processing to be applied to areas other than the face area in the infrared image information (including a rule that no image processing is performed); a face region infrared image processing step (f) for processing a face region in the infrared image information based on the stored face region infrared image processing rule; a non-face area infrared image processing step (g) for processing an area other than the face area in the infrared image information based on the stored non-face area infrared image processing rule; a post-image-processing infrared image information storing step (h) for storing the post-image-processing infrared image information of the non-face region and the post-image-processing infrared image information of the face region in their original positional relationship; an image-processed infrared image information output step (j) for outputting the integrated image-processed infrared image information; An operating program that can be read and executed by an infrared image output device that is a computer having the above configuration.

6. a two-dimensional temperature information storage step (k) of storing the infrared image information detected in the two-dimensional infrared detection step (a) as two-dimensional temperature information, which is information on two-dimensional temperature distribution; an abnormal state determination rule storage step (m) for storing an abnormal state determination rule for determining an abnormal state in the abnormal state information output step described later; an abnormal condition determination step (n) of determining whether or not there is a part with a temperature equal to or higher than a predetermined temperature in the area of ​​the two-dimensional temperature information corresponding to the face area, using a stored abnormal condition determination rule; 6. An operating program that can be read and executed by an infrared image output device that is a computer according to claim 5, further comprising: an abnormal state information output step (p) that outputs abnormal state information when the judgment result in the abnormal state judgment step (n) is that an abnormal state exists.

Images