Human body region detection system, human body region detection method and program

Abstract

To provide a human body area detection system capable of accurately extracting a human body area.SOLUTION: A human body area detection system 10 comprises: a temperature distribution data generation unit 12 for generating temperature distribution data that indicates an indoor temperature distribution; a face area extraction unit 11 for extracting a face area of a human; a temperature calculation unit 13 for calculating a first temperature on the basis of a temperature distribution of the face area in the temperature distribution data, calculating a second temperature on the basis of a temperature distribution of an area except for the face area in the temperature distribution data, and calculating a threshold temperature on the basis of the first and second temperatures; and a human body area extraction unit 14 for extracting, from the temperature distribution data, a human body area on the basis of the first temperature, the second temperature, and the threshold temperature. Therein, the human body area extraction unit 14 extracts, as a human body area, an area in a range in which the temperature is in a range from the highest temperature of the face area to the threshold temperature when the first temperature is higher than the second temperature, meanwhile, extracts, as a human body area, an area in which the temperature is in a range from the threshold temperature to the lowest temperature of the face area when the first temperature is lower than the second temperature.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a human body region detection system, a human body region detection method, and a program for detecting a human body region in a room. [Background technology]

[0002] There is known a technique for setting a temperature close to human body temperature as a threshold for a thermographic image acquired by an infrared sensor, and extracting a portion exceeding the threshold as a human body region (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2017 / 029762 Summary of the Invention [Problem to be solved by the invention]

[0004] When extracting a human body region from a thermographic image, the temperature of the human body is determined to a certain extent, so the threshold temperature for extracting the human body region is also set in advance to a certain temperature range (e.g., 20°C to 40°C). However, if the air temperature falls within the temperature range for extracting the human body region, the background region in the thermographic image will also be extracted as the human body region, making it difficult to extract the human body region. On the other hand, if the temperature range is set too narrow, parts of the human body region, such as clothing, whose temperature is different from that of the human body, will not be extracted as the human body region, and gaps may occur in the region extracted as the human body region. Thus, if the temperature range for extracting the human body region is set too wide, it will be impossible to distinguish between the human body region and the background region, and if the temperature range is set too narrow, it will be impossible to properly extract the human body region, which may result in inaccurate extraction of the human body region.

[0005] The present disclosure provides a human body region detection system and the like that can accurately extract a human body region. [Means for solving the problem]

[0006] a temperature calculation unit that calculates a first temperature based on the temperature distribution of the face region in the temperature distribution data, calculates a second temperature based on the temperature distribution of a region of the temperature distribution data excluding the face region, and calculates a threshold temperature between the first temperature and the second temperature based on the first temperature and the second temperature; and a human body region extraction unit that extracts the human body region from the temperature distribution data based on the first temperature, the second temperature, and the threshold temperature; and when the first temperature is higher than the second temperature, the human body region extraction unit extracts, as the human body region, a region of the temperature distribution data whose temperature is in the range from the highest temperature of the face region to the threshold temperature, and when the first temperature is lower than the second temperature, extracts, as the human body region, a region of the temperature distribution data whose temperature is in the range from the threshold temperature to the lowest temperature of the face region.

[0007] a temperature calculation step of calculating a first temperature based on the temperature distribution of the face region in the temperature distribution data, calculating a second temperature based on the temperature distribution of a region of the temperature distribution data excluding the face region, and calculating a threshold temperature between the first temperature and the second temperature based on the first temperature and the second temperature; and a human body region extraction step of extracting the human body region from the temperature distribution data based on the first temperature and the second temperature, wherein in the human body region extraction step, if the first temperature is higher than the second temperature, a region of the temperature distribution data whose temperature is in the range from the highest temperature of the face region to the threshold temperature is extracted as the human body region, and if the first temperature is lower than the second temperature, a region of the temperature distribution data whose temperature is in the range from the threshold temperature to the lowest temperature of the face region is extracted as the human body region.

[0008] The program in the present disclosure is a program for causing a computer to execute the above-described human body region detection method. [Effects of the Invention]

[0009] According to the human body region detection system and the like of the present disclosure, a human body region can be accurately extracted. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a human body region detection system according to an embodiment. [Figure 2] FIG. 2 is a flowchart showing an example of the operation of the human body region detection system according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of temperature distribution data. [Figure 4] FIG. 4 is a diagram showing an example of the temperature distribution in the face area. [Figure 5]FIG. 5 is a diagram showing an example of the temperature distribution in the region excluding the face region. [Figure 6] FIG. 6 is a histogram showing an example of the temperature distribution in the face region and the temperature distribution in the region excluding the face region. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, more detailed description than necessary may be omitted. For example, detailed description of well-known matters or redundant description of substantially identical configurations may be omitted. This is to avoid unnecessary redundancy in the following description and to facilitate understanding by those skilled in the art.

[0012] The inventors have provided the accompanying drawings and the following description to enable those skilled in the art to fully understand the present disclosure, and do not intend for them to limit the subject matter described in the claims.

[0013] (Embodiment) A human body region detection system according to an embodiment will be described below with reference to FIGS.

[0014] FIG. 1 is a diagram showing an example of the configuration of a human body region detection system 10 according to an embodiment.

[0015] The human body region detection system 10 is a system that detects the human body region of a person indoors, and is installed in, for example, an air conditioner. The human body region detection system 10 is a computer having a processor, memory, etc. The memory is, for example, a ROM (Read Only Memory) and a RAM (Random Access Memory), and can store programs executed by the processor. The human body region detection system 10 includes a face region extraction unit 11, a temperature distribution data generation unit 12, a temperature calculation unit 13, and a human body region extraction unit 14. The face region extraction unit 11, the temperature distribution data generation unit 12, the temperature calculation unit 13, and the human body region extraction unit 14 are realized by, for example, a processor that executes programs stored in memory. The human body region detection system 10 may be a device in which its components are provided in a single housing, or may be a system in which its components are distributed across multiple housings. Although FIG. 1 shows a camera 100 and a thermosensor 200, the camera 100 and the thermosensor 200 may be components of the human body region detection system 10 or may be components outside the human body region detection system 10.

[0016] The camera 100 is a camera that captures images of the interior of a room. The camera 100 is, for example, an RGB camera, and captures images of the interior of a room to obtain an RGB image.

[0017] The thermosensor 200 scans the room to obtain information for generating temperature distribution data (thermoimage) that indicates the temperature distribution in the room.

[0018] The face region extraction unit 11 extracts a human face region. For example, the face region extraction unit 11 extracts a face region based on an image captured by the camera 100. There are no particular limitations on the method for extracting a face region, but for example, a face region can be extracted using Haar-like features or the like.

[0019] The temperature distribution data generation unit 12 generates temperature distribution data that indicates the temperature distribution in a room. For example, the temperature distribution data generation unit 12 generates the temperature distribution data using information obtained from the thermosensor 200. The thermosensor 200 has a horizontal field of view, and the temperature distribution data generation unit 12 can generate temperature distribution data that indicates the two-dimensional temperature distribution of an object present in the space in front of the thermosensor 200. The thermosensor 200 also has a vertical field of view, and can detect the presence of a person in the space in front of the thermosensor 200. The thermosensor 200 has, for example, a group of pixels arranged in a two-dimensional matrix, and the temperature distribution data generation unit 12 can generate two-dimensional temperature distribution data at once. Note that the thermosensor 200 may have a group of pixels arranged one-dimensionally (a line sensor), and the temperature distribution data generation unit 12 may generate two-dimensional temperature distribution data by scanning the group of pixels one-dimensionally. Alternatively, the thermosensor 200 may have one or more pixels, and the temperature distribution data generator 12 may generate two-dimensional temperature distribution data by scanning the one or more pixels two-dimensionally.

[0020] The temperature calculation unit 13 calculates a first temperature based on the temperature distribution of the face area in the temperature distribution data, calculates a second temperature based on the temperature distribution of the area excluding the face area in the temperature distribution data, and calculates a threshold temperature between the first and second temperatures based on the first and second temperatures. Details of the operation of the temperature calculation unit 13 will be described later.

[0021] The human body region extraction unit 14 extracts a human body region from the temperature distribution data based on the first temperature, the second temperature, and the threshold temperature. Details of the operation of the human body region extraction unit 14 will be described later.

[0022] Next, the operation of the human body region detection system 10 will be described with reference to FIG.

[0023] FIG. 2 is a flowchart showing an example of the operation of the human body region detection system 10 according to the embodiment.

[0024] First, the temperature distribution data generating unit 12 generates temperature distribution data (step S11). For example, the temperature distribution data generating unit 12 generates temperature distribution data as shown in FIG.

[0025] Fig. 3 is a diagram showing an example of temperature distribution data. In Fig. 3, the temperature distribution data is represented schematically by shading, with darker colors indicating higher temperatures. Fig. 3 also shows a face region (rectangular frame) superimposed on the temperature distribution data, which will be described later.

[0026] Next, the facial area extraction unit 11 extracts a facial area of ​​a person (step S12). For example, the facial area extraction unit 11 extracts a facial area based on an image captured by the camera 100. For example, in order to superimpose the extracted facial area on the temperature distribution data, the two-dimensional size of the temperature distribution data and the two-dimensional size of the image captured by the camera 100 are matched. For example, if the size of the image captured by the camera 100 is 640 × 480 and the two-dimensional size of the temperature distribution data is 160 × 120, the facial area extraction unit 11 (or the temperature distribution data generation unit 12) reduces (by a factor of four) the coordinate values ​​of the image captured by the camera 100 based on the two-dimensional size of the temperature distribution data, thereby reducing the size of the facial area by a factor of four. This allows the facial area extracted in the image captured by the camera 100 to be superimposed on the temperature distribution data, as shown in FIG. 3.

[0027] The order in which the processing in step S11 and the processing in step S12 are performed may be reversed.

[0028] The facial area extraction unit 11 may extract a facial area based on temperature distribution data. For example, a large amount of temperature distribution data is input, and a trained model that has trained the facial area in each temperature distribution data as correct answer data can be used to extract a facial area based on the temperature distribution data. In this case, the camera 100 does not need to be provided.

[0029] Next, the temperature calculation unit 13 calculates the first temperature and the second temperature (step S13).

[0030] The first temperature is calculated based on the temperature distribution of the face area in the temperature distribution data. Fig. 4 is a diagram showing an example of the temperature distribution of the face area, and the first temperature is calculated from the temperature distribution shown in Fig. 4. For example, the temperature calculation unit 13 may calculate the average temperature of the face area as the first temperature. However, the method of calculating the first temperature is not limited to this.

[0031] The second temperature is calculated based on the temperature distribution of the area excluding the face area in the temperature distribution data. Fig. 5 is a diagram showing an example of the temperature distribution of the area excluding the face area, and the second temperature is calculated from the temperature distribution as shown in Fig. 5. For example, the temperature calculation unit 13 may calculate the temperature of the most frequent value in the temperature distribution of the area excluding the face area as the second temperature. Note that the method of calculating the second temperature is not limited to this.

[0032] Next, the temperature calculation unit 13 calculates a threshold temperature (step S14). The threshold temperature is a temperature between the first temperature and the second temperature, and is calculated based on the first temperature and the second temperature. For example, the temperature calculation unit may calculate a weighted average temperature as the threshold temperature, which is obtained by averaging the first temperature and the second temperature, each multiplied by a weighting coefficient. For example, each weighting coefficient can be calculated using a histogram of the temperature distribution in the face region and the temperature distribution in the region excluding the face region.

[0033] FIG. 6 is a histogram showing an example of the temperature distribution of the face region and the temperature distribution of the region excluding the face region. In FIG. 6, the temperature distribution hatched with lines from the upper left to the lower right is the temperature distribution of the face region, the temperature distribution hatched with lines from the upper right to the lower left is the temperature distribution of the region excluding the face region, and the hatched area with crossed lines is the area where the temperature regions overlap. FIG. 6 also shows a first temperature, which is the average temperature of the face region, and a second temperature, which is the temperature of the most frequent value in the temperature distribution of the region excluding the face region. Note that in FIG. 6, the area around the second temperature (21°C) in the histogram of the temperature distribution of the region excluding the face region is the temperature of the region excluding the human body region (i.e., the background region), and the area around 26°C is the temperature of the region of the human body region that is not exposed by clothing, etc.

[0034] For example, the temperature calculation unit 13 may calculate a variance value of the temperature distribution in the temperature distribution data, and change the weighting coefficient by which each of the first temperature and the second temperature is multiplied based on the variance value. For example, a temperature between the first temperature and the second temperature may be set as a threshold temperature, and inter-class variance may be performed on the higher and lower sides of the threshold temperature, and the weighting coefficient by which each of the first temperature and the second temperature is multiplied may be changed so that the inter-class variance is maximized and the intra-class variance is minimized. This method is effective when setting a threshold value for separating peaks in a histogram.

[0035] For example, Figure 6 shows the threshold temperature when the weighting coefficients multiplied by the first temperature and the second temperature are each 1 / 2. In this case, the threshold temperature can be calculated by (first temperature / 2) + (second temperature / 2).

[0036] The temperature calculation unit 13 may change the weighting coefficients by which the first temperature and the second temperature are multiplied based on the size of the face area. If the size of the face area is large, it is estimated that the human body area in the temperature distribution data is also large, and the weighting coefficients are adjusted according to the size of the human body area. For example, the weighting coefficients by which the first temperature and the second temperature are multiplied may be changed so that the weighting coefficient by which the second temperature is multiplied increases as the size of the face area increases (i.e., so that the threshold temperature approaches the second temperature). This makes it possible to prevent defects in the human body area caused by setting the threshold temperature too close to the first temperature even when the face area (i.e., the human body area) in the temperature distribution data is large.

[0037] Next, the human body region extraction unit 14 extracts a human body region from the temperature distribution data based on the first temperature, the second temperature, and the threshold temperature (step S15). The human body region includes exposed areas such as the face and hands, as well as areas not exposed by clothing, etc. For example, the process in step S15 includes steps S21 to S24.

[0038] The human body region extraction unit 14 determines whether the first temperature is higher than the second temperature (step S21). When the first temperature is higher than the second temperature, this basically means that the air temperature is lower than the body temperature.

[0039] If the first temperature is higher than the second temperature (Yes in step S21), the human body region extraction unit 14 extracts, from the temperature distribution data, a region whose temperature is in the range from the maximum temperature of the face region to the threshold temperature as a human body region (step S22). For example, FIG. 6 shows a temperature range (range from the maximum temperature to the threshold temperature) for extracting a human body region when the first temperature is higher than the second temperature. In addition to the temperature of the exposed face region of a human body region, the temperature of the region not exposed by clothing or the like is also included in this temperature range, and it can be seen that a region not exposed by clothing or the like and having a small temperature difference from the background region (i.e., a region that is difficult to distinguish from the background region) can also be extracted as a human body region. On the other hand, it can be seen that the temperature of the background region is not included in this temperature range, and the background region can be prevented from being extracted as a human body region.

[0040] If the first temperature is not higher than the second temperature (No in step S21), the human body region extraction unit 14 determines whether the first temperature is lower than the second temperature (step S23). When the first temperature is lower than the second temperature, this basically means that the air temperature is higher than the body temperature.

[0041] If the first temperature is lower than the second temperature (Yes in step S23), the human body region extraction unit 14 extracts, from the temperature distribution data, a region whose temperature is in the range from the threshold region to the minimum temperature of the face region as a human body region (step S24). Although the case where the first temperature is lower than the second temperature is not shown, in the same way as when the first temperature is higher than the second temperature, in addition to the exposed face region of the human body region, a region of the human body region that is not exposed due to clothing or the like can be extracted as a human body region, and the background region can be prevented from being extracted as a human body region.

[0042] If the first temperature is not lower than the second temperature (No in step S23), that is, in the special case where the first temperature and the second temperature are the same temperature, the human body region extraction unit 14 cannot extract the human body region and therefore terminates the processing.

[0043] The order in which the process in step S21 and the process in step S23 are performed may be reversed.

[0044] For example, the human body region extraction unit 14 outputs the extracted human body region, and the output human body region is used in many situations, such as for estimating a person's thermal sensation and detecting body temperature.

[0045] The human body region detection system 10 may have a personal recognition function for recognizing people appearing in images captured by the camera 100, and may output a result of personal identification identified by the personal recognition function.

[0046] The human body region detection system 10 may also acquire the indoor temperature and humidity from a temperature and humidity sensor and output the acquired temperature and humidity. The human body region detection system 10 may also estimate the person's thermal sensation from the acquired temperature and humidity and the temperature distribution in the extracted human body region and output the estimated thermal sensation.

[0047] Furthermore, the human body region detection system 10 may acquire air pressure from an air pressure sensor and output the acquired air pressure.

[0048] Furthermore, the human body region detection system 10 may output the image captured by the camera 100, temperature distribution data (thermoimage), average temperature of the face, number of pixels of the face, average temperature of the human body, number of pixels of the human body, and the like.

[0049] As described above, the human body region detection system 10 is a system for detecting a human body region of a person indoors, and includes a temperature distribution data generation unit 12 that generates temperature distribution data indicating the temperature distribution in the room, a face region extraction unit 11 that extracts the person's face region, a temperature calculation unit 13 that calculates a first temperature based on the temperature distribution of the face region in the temperature distribution data, calculates a second temperature based on the temperature distribution of a region excluding the face region in the temperature distribution data, and calculates a threshold temperature between the first and second temperatures based on the first and second temperatures, and a human body region extraction unit 14 that extracts the human body region from the temperature distribution data based on the first temperature, the second temperature, and the threshold temperature. If the first temperature is higher than the second temperature, the human body region extraction unit extracts, as the human body region, a region of the temperature distribution data whose temperature is in the range from the maximum temperature of the face region to the threshold temperature, and if the first temperature is lower than the second temperature, extracts, as the human body region, a region of the temperature distribution data whose temperature is in the range from the threshold temperature to the minimum temperature of the face region.

[0050] According to this method, the temperature range for extracting the human body region is not predetermined as a fixed range, but is automatically set based on a first temperature calculated based on the temperature distribution of the face region and a second temperature calculated based on the temperature distribution of the region excluding the face region. Therefore, if there is a certain difference between the first temperature and the second temperature, the temperature range can be set appropriately, and the human body region can be accurately extracted. Specifically, extraction of the background region as the human body region can be suppressed, and regions that are not exposed by clothing or the like and have a small temperature difference from the background region can be suppressed from being extracted as the human body region, thereby suppressing the occurrence of missing parts in the region extracted as the human body region. Furthermore, when the first temperature is higher than the second temperature, regions whose temperatures are in the range from the maximum temperature of the face region to the threshold temperature are extracted as the human body region. When the first temperature is lower than the second temperature, regions whose temperatures are in the range from the threshold temperature to the minimum temperature of the face region are extracted as the human body region. This allows the human body region to be easily extracted with a small amount of calculation.

[0051] For example, the temperature calculation unit 13 may calculate, as the threshold temperature, a weighted average temperature obtained by averaging the first temperature and the second temperature each multiplied by a weighting coefficient.

[0052] According to this, the threshold temperature is calculated from the weighted average temperature of the first temperature and the second temperature multiplied by the weighting coefficient, so that the temperature range for extracting the human body region can be set more appropriately.

[0053] For example, the temperature calculation unit 13 may calculate a variance value of the temperature distribution in the temperature distribution data, and change the weighting coefficients by which the first temperature and the second temperature are multiplied, based on the variance value.

[0054] This allows appropriate weighting factors to be set based on the variance values, and accordingly, the temperature range for extracting the human body region can be set more appropriately. For example, by setting the weighting factors so that the inter-class variance is maximized and the intra-class variance is minimized, the temperature range for extracting the human body region can be set more appropriately.

[0055] For example, the temperature calculation unit 13 may change the weighting coefficient by which the first temperature and the second temperature are multiplied based on the size of the face area.

[0056] This allows an appropriate weighting factor to be set based on the size of the face region, and accordingly, the temperature range for extracting the human body region can be set more appropriately. For example, by setting each weighting factor so that the larger the size of the face region, the larger the weighting factor by which the second temperature is multiplied becomes, the more appropriately the temperature range for extracting the human body region can be set.

[0057] For example, the temperature calculation unit 13 may calculate the average temperature of the face area as the first temperature.

[0058] In this way, the first temperature may be the average temperature of the face area.

[0059] For example, the temperature calculation unit 13 may calculate the temperature of the most frequent value in the temperature distribution of the area excluding the face area as the second temperature.

[0060] In this way, the second temperature may be the temperature with the highest frequency in the temperature distribution of the area excluding the face area.

[0061] For example, a face area may be extracted based on an image captured by a camera 100 capturing an image of an interior of a room.

[0062] In this way, the face area may be extracted based on the image captured by the camera 100.

[0063] For example, the face area extraction unit 11 may extract the face area based on temperature distribution data.

[0064] In this case, there is no need to provide the camera 100. Furthermore, there is no need to perform processing to superimpose the face area in the image captured by the camera 100 on the temperature distribution data.

[0065] (Other embodiments) As described above, the embodiments have been described as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited to these, and can be applied to embodiments in which appropriate modifications, substitutions, additions, omissions, etc. are made. Furthermore, it is also possible to combine the components described in the above embodiments to create new embodiments.

[0066] For example, in the above embodiment, an example has been described in which the human body region detection system 10 is installed in an air conditioner or the like, but the human body region detection system 10 may also be realized by a server or the like.

[0067] For example, in the above embodiment, the temperature calculation unit 13 calculates the weighted average temperature obtained by averaging the first temperature and the second temperature, each of which has been multiplied by a weighting coefficient, as the threshold temperature. However, this is not limiting. For example, the temperature calculation unit 13 may calculate the average temperature obtained by averaging the first temperature and the second temperature as the threshold temperature.

[0068] For example, the present disclosure can be realized not only as a human body region detection system 10 but also as a human body region detection method including steps (processing) performed by components that make up the human body region detection system 10.

[0069] Specifically, the human body region detection method is a human body region detection method for detecting a human body region in a room, and as shown in FIG. 2, includes a temperature distribution data generation step (step S11) for generating temperature distribution data indicating a temperature distribution in the room, a face region extraction step (step S12) for extracting a human face region, and a temperature calculation step (step S13) for calculating a first temperature based on the temperature distribution in the face region in the temperature distribution data, calculating a second temperature based on the temperature distribution in a region excluding the face region in the temperature distribution data, and calculating a threshold temperature between the first temperature and the second temperature based on the first temperature and the second temperature. and a human body region extraction step (step S15) of extracting a human body region from the temperature distribution data based on the first and second temperatures. In the human body region extraction step, if the first temperature is higher than the second temperature (Yes in step S21), an area of ​​the temperature distribution data whose temperature is in the range from the maximum temperature of the face region to a threshold temperature is extracted as the human body region (step S22), and if the first temperature is lower than the second temperature (Yes in step S23), an area of ​​the temperature distribution data whose temperature is in the range from the threshold temperature to the minimum temperature of the face region is extracted as the human body region (step S24).

[0070] For example, the human body region detection method may be executed by a computer (computer system). The present disclosure may be realized as a program for causing a computer to execute steps included in the human body region detection method. Furthermore, the present disclosure may be realized as a non-transitory computer-readable recording medium, such as a CD-ROM, on which the program is recorded.

[0071] For example, when the present disclosure is realized as a program (software), each step is performed by running the program using hardware resources such as a computer's CPU, memory, input / output circuits, etc. In other words, each step is performed by the CPU acquiring data from memory or input / output circuits, etc., performing calculations on the data, and outputting the calculation results to memory or input / output circuits, etc.

[0072] Furthermore, the components included in the human body region detection system 10 of the above-described embodiment may be realized as an LSI (Large Scale Integration) which is an integrated circuit (IC).

[0073] Furthermore, the integrated circuit is not limited to an LSI, but may be realized by a dedicated circuit or a general-purpose processor. A programmable FPGA or a reconfigurable processor, in which the connections and settings of circuit cells within an LSI can be reconfigured, may also be used.

[0074] Furthermore, if an integrated circuit technology that can replace LSI emerges due to advances in semiconductor technology or other derived technologies, that technology may naturally be used to integrate the components included in the human body region detection system 10 into an integrated circuit.

[0075] As described above, the embodiments have been described as examples of the technology in the present disclosure, and for that purpose, the accompanying drawings and detailed description have been provided.

[0076] Therefore, the components shown in the accompanying drawings and detailed description may include not only essential components for solving the problem, but also non-essential components for solving the problem. Therefore, the fact that these non-essential components are shown in the accompanying drawings or detailed description should not be interpreted as indicating that these non-essential components are essential.

[0077] Furthermore, since the above-described embodiments are intended to illustrate the technology of the present disclosure, various modifications, substitutions, additions, omissions, etc. may be made within the scope of the claims or their equivalents. [Industrial Applicability]

[0078] The present disclosure can be applied to systems that detect people in a room, etc. [Explanation of symbols]

[0079] 10 Human body area detection system 11 Face area extraction unit 12 Temperature distribution data generation unit 13 Temperature calculation section 14 Human body region extraction part 100 cameras 200 Thermosensor

Claims

1. A human body region detection system for detecting a human body region in a room, comprising: a temperature distribution data generating unit that generates temperature distribution data indicating the temperature distribution in the room; a face area extraction unit that extracts a face area of ​​the person; a temperature calculation unit that calculates a first temperature based on a temperature distribution of the face region in the temperature distribution data, calculates a second temperature based on a temperature distribution of a region excluding the face region in the temperature distribution data, and calculates a threshold temperature that is a temperature between the first temperature and the second temperature based on the first temperature and the second temperature; a human body region extraction unit that extracts the human body region from the temperature distribution data based on the first temperature, the second temperature, and the threshold temperature; The human body region extraction unit If the first temperature is higher than the second temperature, an area of ​​the temperature distribution data, whose temperature is in a range from the maximum temperature of the face area to the threshold temperature, is extracted as the human body area; If the first temperature is lower than the second temperature, an area of ​​the temperature distribution data whose temperature is in a range from the threshold temperature to the lowest temperature of the face area is extracted as the human body area. Human body region detection system.

2. the temperature calculation unit calculates, as the threshold temperature, a weighted average temperature obtained by averaging the first temperature and the second temperature each multiplied by a weighting coefficient; The human body region detection system according to claim 1 .

3. the temperature calculation unit calculates a variance value of the temperature distribution in the temperature distribution data, and changes the weighting coefficient by which each of the first temperature and the second temperature is multiplied based on the variance value; The human body region detection system according to claim 2 .

4. the temperature calculation unit changes the weighting coefficient by which the first temperature and the second temperature are multiplied based on the size of the face area; The human body region detection system according to claim 2 .

5. the temperature calculation unit calculates an average temperature of the face area as the first temperature; The human body region detection system according to any one of claims 1 to 4.

6. the temperature calculation unit calculates, as the second temperature, a temperature of a most frequent value in a temperature distribution in the area excluding the face area; The human body region detection system according to any one of claims 1 to 5.

7. the facial area extraction unit extracts the facial area based on an image captured by a camera capturing an image of the interior of the room; The human body region detection system according to any one of claims 1 to 6.

8. the face area extraction unit extracts the face area based on the temperature distribution data; The human body region detection system according to any one of claims 1 to 7.

9. A human body region detection method for detecting a human body region in a room, comprising: a temperature distribution data generating step of generating temperature distribution data indicating the temperature distribution in the room; a face region extraction step of extracting a face region of the person; a temperature calculation step of calculating a first temperature based on a temperature distribution of the face region in the temperature distribution data, calculating a second temperature based on a temperature distribution of a region excluding the face region in the temperature distribution data, and calculating a threshold temperature between the first temperature and the second temperature based on the first temperature and the second temperature; a human body region extraction step of extracting the human body region from the temperature distribution data based on the first temperature and the second temperature, In the human body region extraction step, If the first temperature is higher than the second temperature, an area of ​​the temperature distribution data, whose temperature is in a range from the maximum temperature of the face area to the threshold temperature, is extracted as the human body area; If the first temperature is lower than the second temperature, an area of ​​the temperature distribution data whose temperature is in a range from the threshold temperature to the lowest temperature of the face area is extracted as the human body area. Human body region detection method.

10. A program for causing a computer to execute the human body region detection method according to claim 9.

Images