Human body detection system and program for human body detection system

Abstract

To provide a human body detection system capable of detecting a human body without the use of a thermal imaging camera with a high pixel count, the present invention provides a human body detection system for detecting a human body present within a preset alert range, said system being characterized by comprising: a thermal imaging camera having temperature-measuring elements arranged in an array; and an information processing device that determines whether an object captured by the thermal imaging camera is a human body or not, wherein the width dimension of an area of detection by a single temperature-measuring element at a set distance in the alert range is set to a width equivalent to the thickness of a human body.

Description

Human body detection system, program for human body detection system

[0001] The present invention relates to a human body detection system and a program for the human body detection system.

[0002] A conventional human body detection system is a system described in Prior Art Document 1, which takes an image of a human body in a security area with a thermal imaging camera, processes the thermal image, and issues a warning based on the processing results.

[0003] In order to make it easier to capture the characteristics (shape) of the human body within the surveillance area in such systems, there is a trend toward increasing the resolution of thermal images, that is, increasing the number of pixels in thermal imaging cameras.

[0004] However, increasing the number of pixels may result in an increase in the cost and size of the thermal imaging camera, and in the complexity of image processing.

[0005] JP 2021-163401 A

[0006] The present invention was made in response to the questioning of the conventional technical common sense that increasing the resolution of thermal imaging cameras leads to improved accuracy in human body detection, and aims to provide a human body detection system that can detect human bodies without using a high-pixel thermal imaging camera.

[0007] In other words, the human body detection system of the present invention is a human body detection system that detects human bodies present within a predetermined warning range, and is equipped with a thermal imaging camera having temperature measuring elements arranged in an array, and an information processing device that determines whether an object captured by the thermal imaging camera is a human body, and is characterized in that the width dimension of the detection area by a single temperature measuring element at the set distance of the warning range is set to correspond to the thickness width of a human body.

[0008] The present invention was developed after extensive research by the inventors, who discovered that sufficient human body detection accuracy can be achieved for applications such as alarm systems, even if the width of the detection area of ​​each temperature sensor at the set distance of the surveillance range is increased to the thickness of a human body. In other words, a human body detection system configured in this way can ensure human body detection accuracy within the surveillance range with as few pixels as possible. Therefore, for the same surveillance range, the thermal imaging camera can be made less expensive, more compact, and with simpler image processing than conventional cameras, thanks to the reduced pixel count. For the same number of pixels, the surveillance range can be expanded compared to conventional cameras.

[0009] When a human body is present within the set distance of the warning range, the thickness and width of the human body will fall within the detection range of one or two temperature measuring elements of the thermal imaging camera. Furthermore, a person in a normal posture has a long and slender shape. Therefore, the information processing device is preferably configured to determine a human body when the thermal imaging camera detects a temperature measuring element that detects a temperature that is greater than a predetermined temperature difference from the reference temperature, or a temperature measuring element that detects a temperature within the range of human body temperature, arranged in one or two rows. It is more desirable to determine a human body when three or more temperature measuring elements are arranged in the longitudinal direction.

[0010] The key point here is that the detection of whether or not there is a human body is not determined by detailed shapes as in the past, but by the simple rectangular shape of the arrangement of the temperature measuring elements, and the inventors have discovered that this is sufficient for human body detection.With this configuration, the calculation processing required for human body detection can be simplified, and a human body present within the set distance of the warning range can be accurately detected based on two parameters (the temperature measured by the temperature measuring elements and the arrangement of the measuring elements).

[0011] If a human body is present at a distance closer than the set distance, the thickness of the human body will not fit within the detection range of the two temperature measuring elements of the thermal imaging camera, but will span the detection ranges of three or more temperature measuring elements, resulting in multiple temperature measuring elements detecting the human temperature. If the thermal imaging camera were to determine whether the object captured by the thermal imaging camera is a human body based on all of these multiple temperature measuring elements, the amount of calculation required by the information processing device would increase.

[0012] To solve this problem, when a human temperature detection element that detects a temperature that is greater than or equal to a predetermined temperature difference from a reference temperature or that detects a temperature within the range of human body temperatures is located at the four corners of any rectangle in the thermal imaging camera, the information processing device can regard that rectangular area as a human temperature detection unit area where a human body temperature is detected, and determine that this is a human body based on the arrangement of multiple human temperature detection unit areas.With this configuration, it is possible to determine whether an object captured by the thermal imaging camera is a human body based on the temperature detection elements located at the four corners of each human temperature detection unit area, rather than based on all the temperature detection elements within the multiple human temperature detection unit areas, thereby simplifying the detection calculation process for human bodies located within the warning range.

[0013] The information processing device may perform a smoothing operation such as averaging on the temperatures detected by the temperature measuring elements located at the four corners of any rectangle in the thermal imaging camera, and if the difference between the temperature obtained by the smoothing operation and the reference temperature is greater than a predetermined value, or if the temperature obtained by the smoothing operation is within the range of human body temperatures, the rectangular area may be regarded as a human temperature detection unit area, and may determine that it is a human body based on the arrangement of multiple human temperature detection unit areas. In this way, even if a person's body temperature cannot be detected in part, such as a person carrying a heat source such as a hand warmer or coolant, the smoothing operation can be used to cancel out the effect of the heat source and detect the human body.

[0014] The information processing device can determine that a human body is present when the human temperature detection unit areas are arranged in one or two rows. This not only makes it easier to determine whether a human body is present within the warning range, but also simplifies the calculation process by using the same method as for detecting and determining a human body present at the set distance described above.

[0015] The information processing device may further determine whether or not a human body is present based on temporal changes in the human temperature detection unit area. With this configuration, false detections caused by noise or background influences, such as when the human temperature detection unit area moves very quickly or does not move at all (for example, when there happens to be a heat source such as a boiler emitting human body temperature in the background), can be reduced, and human bodies present within the alert area can be detected with higher accuracy.

[0016] The information processing device can calculate the distance to the human body based on the number of temperature-sensing elements on one side of the human temperature detection unit area. This configuration, for example, can identify the distance between the thermal imaging camera and the human body, allowing for the calculation of position information within the human alert range, i.e., the angle and distance relative to the thermal imaging camera. It also makes it possible to increase the variety of human body detection, such as by changing the alarm sound as a person approaches.

[0017] The information processing device may determine that an object captured by the thermal imaging camera is a non-human body based on the arrangement of the temperature measuring elements that detect the temperature of the non-human body. In this way, it is possible to detect non-human bodies within the surveillance area, and the application to objects other than human bodies can be expanded.

[0018] The information processing device may determine the situation within the alert area based on the relationship between the temperature measuring elements determined to represent a human body and the temperature measuring elements determined to represent a non-human body. Here, the temperature measuring elements determined to represent a human body are the human temperature detection elements or temperature measuring elements constituting the human temperature detection unit area that are the basis for determining that an object captured by the thermal imaging camera is a human body, and the temperature measuring elements determined to represent a non-human body are the temperature measuring elements that detected the temperature of a non-human body that are the basis for determining that an object captured by the thermal imaging camera is a non-human body. In this case, for example, if a human body and a non-human body with a high temperature are detected within the alert area, it can be determined that arson has occurred based on their relative positions, and it can further increase the alert level within the alert area based on this determination.

[0019] The present invention also provides a program for a human body detection system that causes a computer to function as the information processing device, and can achieve the same effects as those described above.

[0020] According to the present invention configured in this way, a human body can be detected without using a high-pixel thermal imaging camera.

[0021] A functional block diagram showing the functions of a human body detection system according to an embodiment of the present invention. A schematic diagram explaining a two-dimensional arrangement of temperature measuring elements of a thermal imaging camera in the same embodiment. A schematic diagram showing a human body detection system in the same embodiment. A flowchart showing a part of the operation of the human body detection system in the same embodiment. An explanatory diagram 1 for explaining the functions of a determination unit in the same embodiment. An explanatory diagram 2 for explaining the functions of a determination unit in each embodiment.

[0022] <First Embodiment> A first embodiment of a human body detection system 100 according to the present invention will be described with reference to the drawings. The present invention detects a human body HB, which is a detection target, existing within a preset warning range A within the imaging range of a thermal imaging camera 2, and notifies the user of the detection result. Specifically, the human body detection system 100 according to this embodiment is used in an alarm system for intruder detection and crime prevention.

[0023] <Configuration of Human Body Detection System 100> As shown in FIG. 1, the human body detection system 100 of this embodiment includes at least a thermal imaging camera 2 and an information processing device 3.

[0024] The thermal imaging camera 2 measures the temperature of an object within the imaging range and outputs the measured temperature as temperature measurement data. As shown in Figure 2, it is equipped with a plurality of temperature measurement elements 21 arranged in a two-dimensional array, optical components (not shown) consisting of lenses, etc., and an output unit (not shown) that processes the output signal of each temperature measurement element 21 and transmits it to the information processing device 3 as temperature measurement data.

[0025] Although a thermopile element is used as the temperature measuring element 21 here, other elements that detect infrared rays, such as a bolometer element or a pyroelectric element, may also be used.

[0026] In this embodiment, the temperature measuring elements 21 are arranged two-dimensionally, for example, in a matrix (8*8 in the figure), and the position of each temperature measuring element 21 is identified by its row and column number. For example, the position of the temperature measuring element 21 arranged in the first row of column a is represented and recognized as array information (a1) on the information processing device 3.

[0027] The temperature measurement data in this embodiment indicates the absolute temperature measured by each temperature measurement element 21 and can be converted to Celsius, Fahrenheit, etc. This temperature measurement data includes arrangement information of the temperature measurement elements 21 that detected that temperature.

[0028] The thermal imaging camera 2 of this embodiment is attached to, for example, a wall with its imaging surface facing horizontally.

[0029] As shown in Figure 3, the imaging range of the thermal imaging camera 2 is determined by the arrangement of the temperature measuring elements 21 and the focal length of the lens (not shown) combined with the thermal imaging camera 2. The viewing angle of a single temperature measuring element 21 is determined by the size of the temperature measuring element 21 and the focal length of the lens. The size of the detection area (hereinafter referred to as unit detection area B) by each temperature measuring element 21 at a location a predetermined distance away from the thermal imaging camera 2 is determined by the predetermined distance, the size of each temperature measuring element 21, and the focal length of the lens of the thermal imaging camera 2.

[0030] 3, the measurement distance at which the dimension of the unit detection area B is equivalent to the thickness width of the human body HB is set as the set distance, and the range within the set distance in the imaging range is set as the security area A. When using a thermal imaging camera 2 with an adjustable optical system, the security area A (and the set distance) can be adjusted by adjusting the optical system.

[0031] The width in the thickness direction of the human body HB is, for example, the thickness of the chest, more specifically, 15 to 30 cm. Note that the width dimension of the human body at the set distance may be equivalent to the horizontal width of the human body HB to be detected (for example, shoulder width or horizontal chest diameter).

[0032] The information processing device 3 is a dedicated or general-purpose computer equipped with a CPU, memory, an input / output interface, an AD converter, etc., and is connected to the thermal imaging camera 2 so as to be able to communicate with it.

[0033] 1, the information processing device 3 functions as at least a storage unit 31, a determination unit 32, and a notification output unit 33. The storage unit 31 is formed in a predetermined area of ​​a memory.

[0034] The storage unit 31 stores reference temperature data indicating a reference temperature for each temperature measuring element 21 and abnormal temperature data indicating an abnormal temperature range for each temperature measuring element 21 .

[0035] The reference temperature data in this embodiment indicates the background temperature of each temperature measuring element 21. The reference temperature data is preferably temperature measurement data measured when no human body HB is present within the surveillance range A of the thermal imaging camera 2, and in this embodiment, it is temperature measurement data measured during calibration performed before the human body detection system 100 is put into operation.

[0036] The reference temperature data may be past temperature measurement data measured during operation of the human body detection system 100, or may be data that is updated during operation. The reference temperature data may also be an average of temperature measurement data over a predetermined period of time, or a moving average calculated from time-series changes in the temperature measurement data.

[0037] The abnormal temperature data indicates a high or low temperature range that would normally not be detected by the human body. For example, the abnormal temperature data is set to a temperature range of 45 degrees or higher.

[0038] The determination unit 32 detects a human body by determining whether or not the object captured by the thermal imaging camera 2 is a human body HB, and outputs the determination result to the notification output unit 33, which will be described later.

[0039] The judgment unit 32 compares the temperature measurement data value with the reference temperature data value for each temperature measuring element 21 and determines whether the difference (temperature difference) is greater than or equal to a predetermined value. The judgment unit 32 also compares the temperature measurement data value of a temperature measuring element 21 whose temperature difference is greater than or equal to a predetermined value with the abnormal temperature data value to determine whether the temperature measuring element 21 is detecting an abnormal temperature. The judgment unit 32 then judges a temperature measuring element 21 whose temperature difference is greater than or equal to a predetermined value but which is not detecting an abnormal temperature as a temperature measuring element 21 that has detected human body temperature (hereinafter referred to as a human temperature detection element X). Note that the judgment unit 32 may also judge a temperature measuring element 21 whose temperature difference is greater than or equal to a predetermined value as a human temperature detection element X.

[0040] The judgment unit 32 judges whether an object present within the surveillance range A of the thermal imaging camera 2 is a human body HB based on the arrangement and position of the human temperature detection elements X on the two-dimensional array of the temperature measuring elements 21.

[0041] In addition, in order to be able to detect a human body HB at both a long distance and a short distance, the judgment unit 32 of this embodiment detects a human body HB when the arrangement or placement of the human temperature detection elements X is in at least one of the following two patterns.

[0042] The first pattern is for long distances, and is a case where the human temperature detection elements X are arranged in only one or two rows in the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2 (Pattern 1). In this embodiment, Pattern 1 is a case where the human temperature detection elements X are arranged in only one or two rows, with three elements arranged consecutively in the longitudinal direction. Pattern 1 may also be a case where four to seven elements X are arranged in the longitudinal direction, and is appropriately set according to the shape of the human body to be detected.

[0043] The second pattern is for short distances, where the human temperature detection elements X are located at the four corners of any rectangle in the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2, and the rectangular area is arranged in only one or two rows (Pattern 2). When Pattern 2 is applied, the judgment unit 32 regards the rectangular area formed by the four human temperature detection elements X as the human temperature detection unit area Y where human body temperature is detected, and determines whether this is a human body HB based on the arrangement of these human temperature detection unit areas Y. The judgment unit 32 determines that a human body HB is present when the human temperature detection unit areas Y are arranged in only one or two rows. Specifically, Pattern 2 is when two to three human temperature detection unit areas Y are arranged consecutively in the longitudinal direction in only one or two rows. The number of human temperature detection unit areas Y that constitute Pattern 2 is determined appropriately depending on the number and arrangement of the temperature measuring elements 21.

[0044] The notification output unit 33 can issue commands to various alarms to issue notifications according to the detection results of the determination unit 32. Examples of the various alarms include warning lights and buzzers. The notification of the detection results may be sent to the detected human body HB or may be sent remotely to the installer of the human body detection system 100.

[0045] The alarm output unit 33 can change the command to be given to the alarm in accordance with the pattern used for human body detection by the determination unit 32. For example, the alarm output unit 33 can command the buzzer to emit a louder sound when a human body HB is detected using short-distance pattern 2 than when a human body HB is detected using long-distance pattern 1.

[0046] <Operation of the human body detection system 100> The operation of the human body detection system 100 of this embodiment when detecting a human body HB will be described below with reference to the flowchart in Fig. 4. Fig. 4 shows a process flow in which the determination unit 32 determines whether an object captured by the thermal imaging camera 2 is a human body HB.

[0047] First, when the human body detection system 100 is activated, the determination unit 32 acquires temperature measurement data measured at a certain timing from each temperature measuring element 21 (S1).

[0048] Based on each temperature measurement data and the reference temperature data acquired from the memory unit 31, the judgment unit 32 judges whether or not there is a temperature measurement element 21 among the temperature measurement elements 21 that has detected a temperature whose temperature difference from the reference temperature is greater than a predetermined value (hereinafter also referred to as an element 21 that has detected a temperature difference) (S2).

[0049] Furthermore, based on the temperature measurement data and the abnormal temperature data acquired from the memory unit 31, the judgment unit 32 judges whether or not there are any temperature measurement elements 21 that have not detected an abnormal temperature among the temperature measurement elements 21 that have detected a temperature whose temperature difference from the reference temperature is greater than or equal to a predetermined value (S3).

[0050] If there is a temperature measuring element 21 that has detected a temperature whose temperature difference from the reference temperature is greater than a predetermined value but has not detected an abnormal temperature, the judgment unit 32 judges this to be a human temperature detecting element X and stores the array information of the human temperature detecting element X (S4).

[0051] Then, based on the stored arrangement information of each human temperature detection element X, the judgment unit 32 judges whether the arrangement and placement of the human temperature detection elements X on the two-dimensional array of the temperature measuring elements 21 matches pattern 1 or pattern 2 (S5).

[0052] If the arrangement and placement of the human temperature detection elements X match pattern 1 or pattern 2, the judgment unit 32 determines that the object present within the surveillance range A of the thermal imaging camera 2 is a human body HB and detects the human body (S6).

[0053] The two patterns of this embodiment described above will be explained in detail with reference to Figure 5. In each of Figures (I) to (VI) in Figure 5, the squares on the right represent the detection status of each of the two-dimensionally arranged temperature measuring elements 21 of the thermal imaging camera 2, and the frame on the left represents an image of the warning area A that the system 100 can derive based on the temperature measurement data of the thermal imaging camera 2. In the squares in Figure 5, the white squares represent temperature measuring elements 21 that are not detecting human body temperature, and the darker squares represent human temperature detecting elements X. The stars on the squares are added for convenience to indicate the temperature measuring elements 21 at the four corners of the human temperature detecting unit area Y.

[0054] 5(I) shows a case where none of the temperature measuring elements 21 of the thermal imaging camera 2 detects a human body temperature. In this case, there is no arrangement of the temperature measuring elements 21 that fits the above pattern in the two-dimensional array of the temperature measuring elements 21, so the judgment unit 32 does not judge that a human body HB exists within the warning area A, or judges that a human body HB does not exist within the warning area A.

[0055] 5(II) shows a case where a cluster of five human temperature detection elements X is arranged in series (b3 to b7) on the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2. This arrangement can be said to be a single row of human temperature detection elements X, and corresponds to pattern 1. In this case, the determination unit 32 determines that a human body HB is present within the alert area A.

[0056] 5(III) shows cases (f2 to f6) where groups of 10 human temperature detection elements X are arranged in series, five in a row, on the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2. This arrangement can be said to be arranged in only two rows of human temperature detection elements X, and corresponds to pattern 1.

[0057] Figure 5 (IV) shows a case where six human temperature detection elements X are arranged in three rows (b3:d8) on the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2. In this figure, a square area (b3:d5) formed by the human temperature detection elements X (b3, d3, b5, d5) and a square area (b6:d8) formed by the human temperature detection elements X (b6, d6, b8, d8) are arranged in series. This arrangement can be said to be a single row of human temperature detection unit areas Y, and corresponds to pattern 2.

[0058] <Effects of this embodiment> As described above, in the human body detection system 100 of the first embodiment, the width dimension of the unit detection area B defined by the single temperature measuring element 21 at the set distance of the surveillance area A is set to be equivalent to the width of the detection target (the thickness width of the human body HB). This ensures the detection accuracy of the human body HB within the surveillance area A with as few pixels as possible. Therefore, if the size of the surveillance area A is the same, the thermal imaging camera 2 can be made less pixelated, resulting in lower costs, smaller size, and simpler image processing compared to conventional cameras. If the number of pixels is the same, the surveillance area A can be expanded compared to conventional cameras.

[0059] In addition, the judgment unit 32 of this embodiment judges whether or not a human body temperature has been detected by the temperature measuring element 21 based on the temperature difference between the measured temperature data and the reference temperature data, so that reference temperature data can be obtained from the environment where the device is installed, eliminating the need to manually create it, which is convenient. Furthermore, since the reference temperature data can be used as the background temperature during calibration, false detections can be reduced and the accuracy of human body detection can be improved when the device is used in a location where the background temperature is close to the human body temperature.

[0060] Furthermore, the determination unit 32 does not use the temperature measuring element 21 that detected an abnormal temperature for human body detection, so that false detections can be reduced using a simple algorithm.

[0061] In this embodiment, when a human body is present within the set distance of the warning range, the determination unit 32 detects the human body HB based on whether the arrangement of the human temperature detection elements X corresponds to pattern 1. This simplifies the calculation process required for human body detection, while enabling accurate detection of a human body present within the set distance of the warning range A based on two parameters (the temperature measured by the temperature measuring elements and the arrangement of the measuring elements).

[0062] In addition, in this embodiment, when the human body HB is present at a distance closer than the set distance, the judgment unit 32 detects the human body HB based on whether the arrangement of the human temperature detection unit areas Y corresponds to pattern 2.

[0063] In this way, it is possible to determine whether an object captured by the thermal imaging camera 2 is a human body HB based on the temperature measuring elements 21 located at the four corners of each human temperature detection unit area Y, rather than based on all the temperature measuring elements 21 within the multiple human temperature detection unit areas Y, thereby simplifying the detection calculation process for a human body HB located within the surveillance area A. Furthermore, not only can it be more easily determined whether a human body HB is located within the surveillance area A, but it is also possible to use a method common to the detection and determination method for a human body HB located at the set distance described above, thereby further simplifying the calculation process.

[0064] Second Embodiment The determination unit in the first embodiment identifies a human temperature detection element based on whether the temperature difference between the temperature measurement data and the reference temperature data is equal to or greater than a predetermined value, that is, based on a relative value. In contrast, the determination unit in the second embodiment identifies a human temperature detection element based on an absolute value. For example, the determination unit in this embodiment determines that a temperature measurement element that measures a temperature within a predetermined range of human body temperatures is a human temperature detection element.

[0065] The predetermined human body temperature is, for example, the temperature that the body temperature (skin temperature) of the human body HB can be. Note that the human-like temperature may also be the temperature that the surface temperature of clothing worn by the human body can be, or the temperature that parts of the human body that are covered other than the skin, such as hair, can be. This human body temperature is set appropriately according to the environment, season, time of day, or detection target (adult / child) of the installation location of the system 100.

[0066] This is useful when detecting a human body HB in an environment with large temperature changes or when detecting a stationary human body HB. Furthermore, the conditions for human body detection can be determined before the system 100 is introduced.

[0067] Third Embodiment A determination unit of a third embodiment detects a human body based on the following two types of patterns in addition to (or instead of) the human body detection based on pattern 1 and pattern 2 described in the first embodiment.

[0068] The third pattern is when the human temperature detection elements X are arranged in one or two rows in the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2 (Pattern 3). The fourth pattern is when the judgment unit 32 determines that the human temperature detection unit areas Y are arranged in one or two rows (Pattern 4).

[0069] Pattern 3 or Pattern 4 will be specifically described with reference to Fig. 6. The format of each diagram in Fig. 6 is the same as the format of each diagram in Fig. 5.

[0070] 6(V) shows a case where there are five human temperature detection elements X arranged consecutively in the column direction on the two-dimensional array of the temperature measuring elements 21 of the thermal imaging camera 2, and one additional human temperature detection element X adjacent to the column of human temperature detection elements X, and the group of human temperature detection elements X is arranged in an L shape. This arrangement does not fall under Pattern 1, as it cannot be said that the human temperature detection elements X are arranged in only one or two columns. On the other hand, since there are at least five human temperature detection elements X arranged in one column (f2 to f6), it falls under Pattern 3.

[0071] 6(VI) shows a case where a large human temperature detection unit area Y (e2:f4) and a small human temperature detection unit area Y (d5:g8) are arranged consecutively. In this case, the two human temperature detection unit areas Y arranged in a vertical line have different widths, so the two human temperature detection unit areas Y cannot be said to be arranged in only one line, and therefore does not fall under Pattern 2. On the other hand, the arrangement of the two human temperature detection unit areas Y (e2:f4) and (d5:g8) is arranged in a vertical line, so it falls under Pattern 4.

[0072] 6 (VII) shows a case where the human temperature detection unit areas (e3:g5) and (d6:f8) arranged in a vertical line are consecutively arranged with a shift in the width direction of the line. In this case, the two human temperature detection unit areas Y are not arranged in a single line, and therefore do not fall under Pattern 2. On the other hand, the two human temperature detection unit areas Y (e3:g5) and (d6:f8) are arranged in a vertical line, so they fall under Pattern 4.

[0073] If a human body is detected based on such patterns 3 and 4, the human body HB can be detected based on a wider variety of arrangements or configurations of the human temperature detection elements X.

[0074] <Fourth embodiment> When detecting a human body based on pattern 2, the determination unit of the first embodiment regards any rectangular area in which human temperature detection elements are located at the four corners as a human temperature detection unit area. In contrast, when detecting a human body based on pattern 2, the determination unit of the fourth embodiment performs a smoothing operation such as averaging on the temperatures detected by the temperature measuring elements located at the four corners of any rectangular area in the two-dimensional array of temperature measuring elements of the thermal imaging camera, and regards the rectangular area as a human temperature detection unit area if the temperature obtained by the smoothing operation has a predetermined temperature difference from a reference temperature or is included in the range of human body temperatures.

[0075] The determination unit of the first embodiment does not regard the rectangular area formed by the four temperature measuring elements (b3, d3, b5, d5) shown in Fig. 6 (VIII) as the human temperature detection unit area Y. In contrast, the determination unit of the fourth embodiment regards the rectangular area formed by the four temperature measuring elements as the human temperature detection unit area Y when, for example, there is a temperature difference of a predetermined value or more between the average of the temperatures detected by the temperature measuring elements (b3, d3, b5, d5) and the average of the reference temperature data of the four temperature measuring elements.

[0076] In this way, even if a person's body temperature cannot be detected in part, such as a person carrying a heat source such as a hand warmer or coolant, the effect can be canceled out by smoothing calculation, making it possible to detect the human body HB.

[0077] <Fifth embodiment> In the judgment unit 32 of the first embodiment, for example, if the human temperature detection unit area Y on the two-dimensional array of temperature measuring elements 21 moves at an inhuman speed or does not move at a speed that is inhuman, there is a possibility that it will be mistakenly detected as a human body HB.

[0078] Therefore, the determination unit of the fifth embodiment determines whether or not an object is a human body based on the temporal change of the human temperature detection elements or the human temperature detection unit area. For example, if the amount of change per hour in the arrangement of the human temperature detection elements is equal to or greater than a predetermined amount, the determination unit determines that the object captured by the thermal imaging camera 2 is a non-human body. If the surveillance area is an area where people pass by, the predetermined amount of change is set based on the movement (walking) speed of the person.

[0079] The determination unit of the fifth embodiment detects the direction of movement of a human body within the alert area based further on the human temperature detection element or the temporal change in the human temperature detection unit area.

[0080] The determination unit of the fifth embodiment can detect a human body HB based on temporal changes in the human temperature detection element or human temperature detection unit area, so that in cases where the human temperature detection element or human temperature detection unit area moves very quickly or does not move at all (for example, when there happens to be a heat source such as a boiler that emits human body temperature in the background), false detections due to the influence of noise or background, etc., can be reduced, and a human body present within the alert area can be detected with higher accuracy. Furthermore, the determination unit of this embodiment can detect the direction of movement of a human body, which is useful for managing the flow of people within a facility.

[0081] Sixth Embodiment The information processing device of the sixth embodiment functions as a distance calculation unit that calculates the distance to a detected human body. When the determination unit detects a human body in pattern 1 (or 3), the distance calculation unit calculates the set distance of the thermal imaging camera's surveillance range as the distance between the detected human body and the thermal imaging camera. When the determination unit detects a human body in pattern 2 (or 4), the distance calculation unit calculates the distance between the detected human body and the thermal imaging camera based on the number of temperature measuring elements on one side of the human temperature detection unit area and the width dimension of the unit detection area defined by a single temperature measuring element at the set distance.

[0082] The information processing device 3 of the sixth embodiment functions as a distance calculation unit, and can, for example, determine the distance between the thermal imaging camera and a human body, thereby calculating the position information of the human body within the surveillance range, i.e., the angle and distance based on the thermal imaging camera. In addition, it is possible to increase the variety of human body detection, such as changing the alarm sound as the human body approaches.

[0083] Seventh Embodiment An information processing device according to a seventh embodiment functions as a non-human body determining unit that identifies detection targets other than human bodies, and as a situation determining unit that determines the situation within the alert area.

[0084] The non-human body determination unit determines whether an object captured by the thermal imaging camera is a non-human body based on the arrangement of the temperature measuring elements (hereinafter referred to as non-human body temperature detecting elements) that detected the temperature of the non-human body. For example, if a vehicle is to be detected as the object, reference temperature data (car temperature data) for detecting cars is stored in advance in the memory unit. The non-human body determination unit identifies the temperature measuring element (non-human body temperature detecting element) that detected the temperature of the car based on the car temperature data and the temperature measurement data, and determines whether the object captured by the thermal imaging camera is a car based on the arrangement of the temperature measuring elements.

[0085] The situation determination unit identifies a situation within the alert area based on the positional relationship between the human temperature detection element or human temperature detection unit area that is the basis for the determination unit's detection of a human body and the non-human body temperature detection element that is the basis for the non-human body determination unit's detection of a non-human body. For example, if there is a non-human body temperature detection element detecting a high temperature in the immediate vicinity of a human temperature detection element, the situation determination unit identifies the situation as dangerous.

[0086] The situation determination unit also has the function of determining that an abnormal situation has occurred if any of the non-human body temperature detection elements measures a particularly abnormal temperature. For example, if any of the temperature measurement elements detects a non-human body temperature of 60 degrees or higher, the situation determination unit determines that an abnormal heat generation or fire has occurred.

[0087] The information processing device of the seventh embodiment functions as a non-human body determining unit, and therefore can be applied to objects other than human bodies present within the surveillance area, thereby expanding the range of uses.

[0088] The situation assessment unit of the seventh embodiment can identify the situation within the alert area based on the human body detection element and the non-human body temperature detection element, allowing the human body detection system to perform a variety of functions tailored to the situation. For example, if a human body and a non-human body with a high temperature are detected within the alert area, it can determine that a fire has been set based on their relative positions, and the alert level within the alert area can be increased based on this determination. Furthermore, since the situation assessment unit can detect the occurrence of an abnormal situation based on the non-human body temperature detection element, the human body detection system can be given the functionality of, for example, a fire detector. It can also be used to detect abnormal heat generation from materials at construction sites.

[0089] <Other Embodiments> As shown in FIG. 6 (IX), the human body temperature detection unit area may be a rectangular area formed by human body temperature detection elements positioned inside a cluster of human body temperature elements.

[0090] In the above embodiments, the shape of the human temperature detection unit area is an arbitrary rectangle, but this may also be a specific rectangle. For example, the shape of the human temperature detection unit area may be specified as a square, the length of each side of which is the length of two or three temperature measuring elements lined up. Multiple types of this specific rectangle may be set.

[0091] In the above embodiments, the human temperature detection unit area was an arbitrary rectangular area formed by four human temperature detection elements. However, the human temperature detection unit area may be, for example, an arbitrary polygonal area formed by three or more human temperature detection elements. For example, if the human temperature detection elements are located at the three corners of an arbitrary triangle, the triangle may be regarded as the human temperature detection unit area. This also simplifies the human body detection calculation process compared to detecting a human body based on all the temperature measuring elements within the human temperature detection unit area.

[0092] In the above embodiments, the determination unit considers a rectangular area as a human temperature detection unit area if a human temperature detection element is located at one of the four corners of the rectangle. However, the determination unit may also consider a rectangular area as a human temperature detection unit area if a human temperature detection element is located at one of the four corners of the rectangle and if another human temperature detection element is located at a predetermined position (e.g., near the center) within the rectangular area. The determination unit may also perform a smoothing operation on the temperatures detected by the temperature measurement elements located at the four corners of the rectangular area and the temperature measurement elements located at a predetermined position within the rectangular area, and consider the rectangular area as a human temperature detection unit area if the temperature obtained by the smoothing operation is within a predetermined temperature range from a reference temperature or is within the range of human body temperatures. There may be multiple predetermined positions within the rectangular area. This simplifies the human body detection calculation process while improving the accuracy of human body detection.

[0093] When detecting a human body based on the arrangement of the human temperature detection unit areas, the determination unit may not use rectangular areas containing temperature measuring elements that detected abnormal temperatures for its determination. In this case, the determination unit does not use rectangular areas containing temperature measuring elements that detected abnormal temperatures for human body detection, so it will not determine that a human body is present in rectangular areas containing temperature measuring elements that measured temperatures that are impossible for a human body, thereby improving the detection performance of the human body detection system.

[0094] The determination unit that detects a human body may determine whether the detected human body is a human body with an abnormal body temperature (hyperthermia or hypothermia) based on the temperature measurement data of the human temperature detection elements or the arrangement and placement of the human temperature detection elements. The determination unit that detects a human body may also determine the size of the detected human body, for example, whether it is an adult or a child, based on the temperature measurement data of the human temperature detection elements or the arrangement and placement of the human temperature detection elements. This is convenient for managing admission to facilities, etc.

[0095] Although the thermal imaging camera 2 in the above embodiment is mounted on a wall, the thermal imaging camera may be mounted on a ceiling or the like with its imaging surface facing vertically. When the thermal imaging camera is mounted on a ceiling, it can detect a human body HB as shown in Figure 6 (IX). If the set distance of the thermal imaging camera's surveillance range is adjusted to approximately the distance between the ceiling and the ground, the thermal imaging camera can detect a human body HB with an arrangement of two or three human temperature detection elements.

[0096] Although the number of pixels of the thermal imaging camera 2 in the above embodiment is 64, the number of pixels of the thermal imaging camera is not limited to this. A thermal imaging camera with a higher number of pixels may be used, in which case the set distance can be set to a long distance, allowing for detection of a human body at a long distance. If a thermal imaging camera with a lower number of pixels is used, the human body detection system can be configured more simply.

[0097] The human body detection system of each of the above embodiments may detect human bodies present in the surveillance area of ​​the thermal imaging camera in real time while measuring the temperature with the thermal imaging camera 2, or may detect human bodies present in the surveillance area A of the thermal imaging camera based on temperature measurement data obtained in advance.

[0098] In the above embodiments, the determination unit detects a human body when the human temperature detection elements or human temperature detection unit areas are arranged in one or two rows, but the determination unit may also detect a human body when the human temperature detection elements or human temperature detection unit areas are arranged in a specific pattern. The specific pattern may be, for example, a dogleg, 2*1, 3*1, 4*2, or multiple rows of two or more.

[0099] The determination unit 32 in each of the above embodiments detects a human body when a plurality of human temperature detection unit areas are arranged, but may detect a human body based on a single human temperature detection unit area.

[0100] While the human body detection system according to the embodiment described above uses one thermal imaging camera, the human body detection system according to the present invention may be configured with multiple thermal imaging cameras. This allows for a wider surveillance range. Using multiple low-pixel thermal imaging cameras can significantly reduce costs and space requirements for a larger surveillance range. Furthermore, according to the present invention, the amount of information that the information processing device must process to detect a human body is small, so even if the number of thermal imaging cameras is increased, the amount of information to be processed does not become too large.

[0101] For example, by combining two or four thermal imaging cameras with a 90-degree angle of view, a human body detection system can be constructed with a surveillance area A with a 180-degree or 360-degree angle of view.

[0102] The human body detection system of the above embodiment may be combined with another PIR sensor, etc., and configured to switch on and off temperature detection by the thermal imaging camera 2 based on the output of the PIR sensor. By utilizing this, standby mode, alert mode, etc. can be set, thereby saving power.

[0103] The present invention also includes a program for a human body detection system that causes a computer to function as the information processing device 3 of each of the above embodiments. By using such a program, it is possible to detect human bodies using an existing thermal imaging camera while simplifying the calculation processing of the information processing device.

[0104] According to the present invention, the accuracy of detecting a human body within the surveillance area can be ensured with as few pixels as possible. Therefore, if the surveillance area is the same, the thermal imaging camera can be made less pixelated, which reduces costs, makes it smaller, and simplifies image processing compared to conventional cameras. If the number of pixels is the same, the surveillance area can be expanded compared to conventional cameras.

[0105] REFERENCE SIGNS LIST 100: Human body detection system 2: Thermal imaging camera 21: Temperature measuring element X: Human temperature detection element Y: Human temperature detection unit area 3: Information processing device 31: Storage unit 32: Determination unit 33: Notification output unit A: Warning range B: Unit detection area HB: Human body

Claims

1. A human body detection system that detects a human body within a pre-set warning range, comprising: a thermal imaging camera having temperature measuring elements arranged in an array; and an information processing device that determines whether an object captured by the thermal imaging camera is a human body, wherein the width dimension of the detection area detected by a single temperature measuring element at the set distance of the warning range is set to correspond to the thickness width of a human body.

2. The human body detection system of claim 1, wherein the information processing device determines that a human body is present when the temperature measuring elements in the thermal imaging camera detect a temperature whose temperature difference from a reference temperature is greater than or equal to a predetermined value, or the human temperature detecting elements detect a temperature that is within the range of human body temperatures, and are arranged in one or two rows.

3. The human body detection system of claim 1, wherein when a human temperature detection element, which is a temperature measuring element in the thermal imaging camera that detects a temperature whose temperature difference from a reference temperature is greater than or equal to a predetermined value, or a human temperature detection element that is a temperature measuring element that detects a temperature included in the range of human body temperatures, is located at the four corners of any rectangle, the information processing device regards the rectangular area as a human temperature detection unit area in which human body temperature is detected, and determines that this is a human body based on the arrangement of multiple human temperature detection unit areas.

4. The human body detection system of claim 1, wherein the information processing device performs a smoothing operation such as averaging on the temperatures detected by the temperature measuring elements located at the four corners of any rectangle in the thermal imaging camera, and when the temperature difference between the temperature obtained by the smoothing operation and a reference temperature is greater than a predetermined value, or when the temperature obtained by the smoothing operation is included in the range of human body temperatures, the information processing device regards the rectangular area as a human temperature detection unit area and determines that it is a human body based on the arrangement of multiple human temperature detection unit areas.

5. A human body detection system according to claim 3 or 4, wherein the information processing device determines that a human body is present when the human temperature detection unit areas are arranged in one or two rows.

6. A human body detection system according to claim 3 or 4, wherein the information processing device determines whether or not there is a human body based further on a change over time in the human temperature detection unit area.

7. A human body detection system according to claim 3 or 4, wherein the information processing device calculates the distance to the human body based on the number of temperature measuring elements on one side of the human temperature detection unit area.

8. A human body detection system as described in claim 2, 3 or 4, wherein the information processing device determines that an object captured by the thermal imaging camera is a non-human body based on the arrangement of temperature measuring elements in the thermal imaging camera that detect the temperature of the non-human body.

9. A human body detection system according to claim 8, wherein said information processing device determines the situation within said alert area based on the relationship between temperature measuring elements judged to represent a human body and temperature measuring elements judged to represent a non-human body.

10. A program for a human body detection system, which causes a computer to function as the information processing device according to claim 2.

Images