Thermal Image Localization Method and System

Abstract

A thermal image positioning method for determining the position of a heat source in a thermal image, the method comprising obtaining a temperature array corresponding to the thermal image and determining a region of interest in the temperature array. The thermal image positioning method further comprises determining a center reference point of the region of interest, determining a plurality of corner points corresponding to the region of interest, and determining the position of the heat source based on at least one of the plurality of corner points. A thermal image positioning system is also disclosed herein.

Description

Technical Field

[0001] The present invention relates to a thermal image positioning method and system, and more particularly to a thermal image positioning method and system capable of accurately positioning the position of a heat source in a thermal image. Background Art

[0002] Recently, thermal imaging devices have been widely used to detect the body temperature of humans to quickly and efficiently determine whether the body temperature of a passing person is too high. In the existing thermal imaging technology field, a specified temperature threshold is usually used as the judgment basis for the human body. It is assumed that if the highest temperature point in the temperature array region exceeds the temperature threshold, it is defined as the forehead position, and the temperature at the forehead position is used as the body temperature of the human body. However, this method often results in a high misjudgment rate.

[0003] In addition, at different temperatures or environments, the highest temperature point of the human body may not be on the forehead of the human body, or the positioning of the highest temperature point falls on parts such as the nose, mouth, and neck. In practical applications, the existing thermal imaging devices cannot accurately locate the forehead position of the human body. To this end, one of the existing solutions is to be paired with a visible light camera to detect and locate the face position of the human body through image use. Since this method requires an additional camera module and consumes high computing resources. Therefore, there is a need to improve the existing thermal imaging technology for detecting human body temperature. Summary of the Invention

[0004] In view of this, the present invention provides a thermal image positioning method and system, which determine a human body region and a position of the highest temperature point through thermal image data to accurately locate a forehead position.

[0005] An embodiment of the present invention discloses a thermal image positioning method for determining the position of a heat source in a thermal image, which includes obtaining a temperature array corresponding to the thermal image, determining a region of interest (ROI) in the temperature array, determining a center reference point of the region of interest, determining a plurality of corner points corresponding to the region of interest, and determining the heat source position according to at least one of the plurality of corner points.

[0006] Another embodiment of the present invention discloses a thermal image positioning system for determining the position of a heat source in a thermal image, which includes a thermal imaging device and a processing device. The thermal imaging device is used to generate a temperature array corresponding to the thermal image. The processing device is coupled to the thermal imaging device and is used to determine a region of interest in the temperature array, determine a center reference point of the region of interest, determine a plurality of corner points corresponding to the center reference point of the region of interest, and determine the heat source position according to at least one of the plurality of corner points. Brief Description of the Drawings

[0007] Figure 1 Schematic diagram of a thermal image positioning system according to one embodiment of the present invention.

[0008] Figure 2 Schematic diagram of a thermal image positioning process according to one embodiment of the present invention.

[0009] Figure 3A Schematic diagram of a temperature array according to one embodiment of the present invention.

[0010] Figure 3B Schematic diagram of a regional broken temperature array according to one embodiment of the present invention.

[0011] Figure 4 Schematic diagram of determining a region of interest according to one embodiment of the present invention.

[0012] Figure 5 and Figure 6 Schematic diagram of the corresponding relationship between the corner points of the region of interest and a human face according to one embodiment of the present invention.

[0013] Figure 7 Schematic diagram of a thermal image positioning process according to one embodiment of the present invention. Detailed implementation manners

[0014] Please refer to Figure 1 , Figure 1 Schematic diagram of a thermal image positioning system 10 according to one embodiment of the present invention. The thermal image positioning system 10 is used to determine the position of a heat source in a thermal image. The thermal image positioning system 10 includes a thermal imaging device 102 and a processing device 104. The thermal imaging device 102 includes a sensor 1022. The sensor 1022 can be a temperature sensor (not shown in the figure) for sensing a temperature, but is not limited thereto. The thermal imaging device 102 is used to generate a thermal image according to the temperature sensed by the sensor 1022. In addition, the thermal imaging device 102 is further used to generate a temperature array corresponding to the thermal image.

[0015] For example, the thermal imaging device 102 may include a thermal image capture device (not shown in the figure). The thermal image capture device can generate a thermal image and a corresponding temperature array according to the temperature sensed by the sensor. Each array element of the temperature array can correspond to each pixel of the thermal image. The processing device 104 is coupled to the thermal imaging device 102 to determine a region of interest (ROI) ROI in the temperature array and determine a reference point P of the center of the region of interest ROI ROI, and determine multiple corner points corresponding to the region of interest (ROI), and determine the heat source position according to at least one of the corner points. For example, the processing device 104 can be a device including an arithmetic unit for processing the temperature array and the thermal image from the thermal imaging device 102. In this way, the thermal image positioning system 10 according to an embodiment of the present invention can determine the corresponding region of interest (ROI) and the center reference point P of the region of interest through the temperature array. ROI , to accurately locate the position of a forehead of a subject.

[0016] Specifically, please refer to Figure 2 , Figure 2 , which is a schematic diagram of a thermal image positioning process 20 according to an embodiment of the present invention. The thermal image positioning process 20 can be used in the thermal image positioning system 10 and includes the following steps:

[0017] Step 202: Start.

[0018] Step 204: Obtain a temperature array corresponding to the thermal image.

[0019] Step 206: Determine the region of interest (ROI) in the temperature array.

[0020] Step 208: Determine the center reference point P of the region of interest (ROI). ROI .

[0021] Step 210: Determine the corner points corresponding to the region of interest (ROI).

[0022] Step 212: Determine the heat source position according to at least one of the corner points.

[0023] Step 214: End.

[0024] According to the thermal image positioning process 20, in step 204, the thermal imaging device 102 starts to obtain thermal image data corresponding to the thermal image through the sensor 1022 and transmits it to the processing device 104. In an embodiment, the thermal image data can be an M T *M T temperature array, such as an 80*80 temperature array, but not limited thereto.

[0025] To locate the position of the forehead of the subject in the thermal image, in step 206, the processing device 104 may determine the region of interest ROI in the temperature array. Since the thermal image may contain an object, such as coffee or a hot drink, within a temperature range higher than the human body temperature range, which affects the accuracy of locating the heat source position, it is necessary to first filter out the array elements in the temperature array that are higher than a temperature threshold T. Specifically, using the temperature threshold T as a threshold, the array elements with the highest temperature value in the temperature array and the array elements adjacent to them are outlined, and the values of all the outlined array elements are set to negative values, so that the temperature values of the adjusted array elements are less than the temperature threshold T, thereby avoiding affecting the data of the entire temperature array to obtain the temperature array corresponding to the thermal image.

[0026] Specifically, please refer to Figure 3A , Figure 3A which is a schematic diagram of a temperature array M according to an embodiment of the present invention. In Figure 3A , an N T *N T array mask is used to identify the array element with the highest temperature value in the temperature array M to outline the array elements in the temperature array M that are higher than the temperature threshold T. The temperature threshold T may be a value higher than the average human body temperature, such as above 42.5 degrees Celsius, which is not the average temperature of the human body, but is not limited thereto. For example, Figure 3A the array elements A and B in

[0027] are both 45 degrees Celsius, so it is necessary to further determine that the objects located at array elements A and B are not the subject. Therefore, the array elements surrounding array elements A and B are outlined in this step, and their corresponding array elements are also outlined, and the values of array elements A and B and the values of their surrounding array elements are also set to negative values, thereby filtering out the high-temperature value array elements A and B and their surrounding array elements. When the array elements A and B and their surrounding array elements are set to negative values, it means that the temperature array M has been adjusted, and the adjusted temperature array M will not show the temperature of non-subject objects. Figure 3B , Figure 3B which is a schematic diagram of the regional fragmentation temperature array according to an embodiment of the present invention. In Figure 3B , the values of array elements A and B are adjusted from 45 degrees to -45 degrees, and the array elements surrounding array elements A and B are also adjusted to negative values in this step. That is, the processing device 104 uses a 3*4 array mask and a 4*6 array mask to identify the array elements A and B with the highest temperature value and their corresponding array elements in the temperature array M, and outlines the corresponding regional fragmentation temperature arrays FA_A and FA_B, that is, the regional fragmentation temperature arrays FA_A and FA_B contain the array elements A and B that are higher than the temperature threshold T and their surrounding array elements.

[0028] Similarly, as shown in Figure 3B , assuming that the temperature threshold T is set to 39 degrees Celsius, the array elements in the temperature array M whose values exceed 39 degrees and their respective surrounding array elements are also framed, and the values of their corresponding array elements will be set to negative values, that is, the array elements in the filtered area broken temperature arrays FA_A and FA_B are filtered to avoid affecting the data of the entire temperature array M. Thus, Figure 3B the area broken temperature array FA_A of the temperature array M shown in is a 3*4 array mask, and the area broken temperature array FA_B is a 4*6 array mask. It should be noted that the numerical values in the above embodiments are only used for illustration and are not used to limit the present invention.

[0029] Next, in order to find the position of the area of the human face in the thermal image, in step 206, the processing device 104 can further use the adjusted M T *M T to determine a high-temperature area HTZ in the temperature array M, and determine a center point Pcenter, a highest temperature point Pmax in the area, and an average temperature Tavg in the area of the high-temperature area HTZ, where the center point Pcenter is a position of the center point of the high-temperature area HTZ, the highest temperature point Pmax in the area is a position with the highest temperature in the high-temperature area HTZ, and the average temperature Tavg in the area is the average value of the values in all the temperature arrays M in the high-temperature area HTZ. Therefore, the processing device 104 can use the center point Pcenter, the highest temperature point Pmax in the area, and the average temperature Tavg of the high-temperature area HTZ to R *N R array mask to determine the region of interest ROI in the temperature array M. It should be noted that the R *N R array mask is not necessarily equal to the range included in the high-temperature area HTZ.

[0030] Specifically, please refer to Figure 4 , Figure 4 which is a schematic diagram for determining the region of interest ROI in an embodiment of the present invention. In Figure 4 , assuming that the coordinates of an origin O in the thermal image are (0, 0), where the values on the X-axis increase to the right and the values on the Y-axis increase downward, the R *N R array mask can be a 7*7 array mask, used to determine the high-temperature area HTZ in the temperature array M corresponding to the thermal image, and determine the center point Pcenter, the highest temperature point Pmax in the area, and the average temperature Tavg in the area in the high-temperature area HTZ. For example, Figure 4The high-temperature region HTZ in it can be shielded by a 5*5 array, but not limited thereto, N R *N R The array shield can be larger than the array shield of the high-temperature region HTZ.

[0031] In one embodiment, in step 208, the processing device 104 determines, according to an Ns*Ns array shield, from the center point Pcenter of the high-temperature region HTZ to N R *N R Four corners of the array shield respectively determine a closest region, and in step 210, respectively determine the corner points corresponding to the four corners of the N R *N R array shield of the closest region. For example, in Figure 4 the embodiment, the Ns*Ns array shield can be a 3*3 array shield, and with the center point Pcenter as the center, respectively determine the closest regions A1 to A4 from the center point Pcenter to the four corners of the 7*7 array shield. It should be noted that the N R *N R array shield is larger than the Ns*Ns array shield, and the numerical values in the above embodiments are only for illustration and not for limiting the present invention.

[0032] Taking the closest region A1 as an example, the closest region A1 includes array elements A1_1 to A1_9. The processing device 104 respectively compares the numerical values of the array elements A1_1 to A1_9 with the average temperature Tavg in the region. Assuming that the comparison results show that the numerical values of the array elements A1_5, A1_6, A1_8, and A1_9 are higher than the average temperature Tavg in the region, it means that the array elements A1_5, A1_6, A1_8, and A1_9 belong to an inner corner region, while the array elements A1_1 to 4 and A1_7 belong to an outer corner region. Therefore, it can be determined that the closest region A1 has a corner point P1; similarly, for the closest region A2, the array elements A2_4, A2_5, A2_7, and A2_8 belong to the inner corner region, while the array elements A2_1 to 3, A2_6, and A2_9 belong to the outer corner region. Therefore, it can be determined that the closest region A2 has a corner point P2. And so on, the closest region A3 has a corner point P3 and the closest region A4 has a corner point P4.

[0033] Further, the processing device 104 can use the corner point P1 as a reference, and determine the center reference point P of the region of interest according to an upper side length Wtop determined by the above corner points P1 and P2 and a side length H determined by the corner points P1 and P3 or P2 and P4 ROIFor example, the side length H can be a left side length Hleft determined by the corner points P1 and P3. That is, when the coordinate position of the corner point P1 is (P1_x, P1_y), the center reference point P of the region of interest ROI has a coordinate position of (P1_x + Wtop / 2, P1_y + Hleft / 2). In other words, the side length H can also be a right side length Hright determined by the corner points P2 and P4. When the coordinate position of the corner point P2 is (P2_x, P2_y), the center reference point P of the region of interest ROI has a coordinate position of (P2_x + Wtop / 2, P2_y + Hright / 2).

[0034] Since the hottest points on the upper half of the human body may fall on the forehead, nose, mouth or neck of the face, that is, the distribution positions of the hottest points in the thermal image will generally fall near the same vertical line. Therefore, the position of the hottest point Pmax in the region determined by the processing device 104 should fall near or on one of the above-mentioned hottest points.

[0035] Please refer to Figure 5 , Figure 5 , which is a schematic diagram of the correspondence between the corner points P1 - P4 of the region of interest ROI of the embodiment of the present invention and the human face. In the thermal image positioning process 20 of one embodiment of the present invention, first, the region of interest ROI corresponding to the high-temperature region (i.e., the region of the human head) is determined from the temperature array M of the thermal image. Then, in step 212, the processing device 104 determines the heat source position based on at least one of the corner points P1 - P4 of the region of interest ROI.

[0036] In Figure 5 , the processing device 104 first defines a center point interval Icenter to determine whether the center reference point P of the region of interest ROI falls within the center point interval Icenter to determine whether the range of the region of interest ROI is correct (i.e., the region of the human head). When the center reference point P of the region of interest ROI falls within the center point interval Icenter, the processing device 104 determines the heat source position based on at least one of the corner points of the N R *N R array mask. It should be noted that the center point interval Icenter is generated based on the hottest point Pmax in the high-temperature region HTZ and the N R *N R array mask. For example, the center point interval Icenter can be a multiple of the side length N R *N R of the N R array mask (such as 1*N R 、2*NR 、 3*N R etc.), and the side length N R is approximately less than the width of a human face, but not limited thereto.

[0037] Regarding N R *N R At least one of the multiple corner points of the array mask is used to determine the heat source position. In one embodiment, assuming that the center point interval Icenter is 1*N R , and when the region of interest ROI is the human head, the heat source position P is determined by the corner points P1, P2 and the following formula (1) source (i.e., the forehead position). P source (x,y) = (Pn_x, Pn_y + 1), n = 1, 2 (if P1_y > P2_y, n = 1, else n = 2)...(1)

[0038] Formula (1) is used to determine whether the center reference point P of the region of interest ROI falls within the center point interval Icenter where the highest temperature point Pmax in the region is located. Therefore, when the difference between the X-axis coordinate of the center reference point P of the region of interest ROI and the X-axis coordinate of the highest temperature point Pmax in the region is less than 1 / 2*N R (i.e., less than 1 / 2*N R pixels), then the heat source position P source (x,y) = (Pn_x, Pn_y + 1). It should be noted that formula (1) judges the Y-axis coordinates of the corner points P1, P2 and finds the one closer to the upper part of the thermal image. Therefore, the one with the larger Y-axis value is used as the heat source position. In addition, according to the results in actual applications, the position of the forehead usually falls below the corner points. Therefore, the Y-axis coordinate of the heat source position P source (x,y) usually adds a value of 1 and shifts downward in the thermal image, but the shifted value is not limited to this embodiment.

[0039] In one embodiment, when the region of the region of interest ROI is deformed or shifted, the corresponding relationship between the corner points P1 to P4 of the region of interest ROI and the human face may be as Figure 6 shown. In Figure 6 this embodiment, the corner points P1, P3, P4 fall on the human body B_1, while the corner point P2 is located on the human body B_2. The corner point P2 is located on another human body. In this case, the center reference point P of the region of interest ROI cannot meet the conditions of formula (1). Therefore, the processing device 104 needs to further determine the heat source position P by the corner points P1, P2 and the following formula (2) or formula (3) source (i.e., the forehead position).

[0040] (|P ROI _x - Pmax_x| > 1 / 2 * N R ) & (|P1_x - Pmax_x| < 1 / 2 * N R )

[0041] →P source (x, y) = (P1_x, P1_y + 1)...(2)

[0042] (|P ROI _x - Pmax_x| > 1 / 2 * N R ) & (|P2_x - Pmax_x| < 1 / 2 * N R )

[0043] →P source (x, y) = (P2_x, P2_y + 1)...(3)

[0044] Equations (2) and (3) are used to determine whether the difference between the X-axis coordinate of corner point P1 or corner point P2 and the X-axis coordinate of the highest temperature point Pmax in the region is less than 1 / 2 * N R (that is, less than 1 / 2 * N R pixels). Specifically, for Equation (2), when the X-axis of corner point P1 and the X-axis coordinate of the highest temperature point Pmax in the region are less than 1 / 2 * N R , then the heat source position P source = (P1_x, P1_y + 1); for Equation (3), when the X-axis of corner point P2 and the X-axis coordinate of the highest temperature point Pmax in the region are less than 1 / 2 * N R , then the heat source position P source = (P2_x, P2_y + 1).

[0045] In another case, when both corner points P1 and P2 fall outside the center point interval Icenter, that is, when both corner points P1 and P2 cannot satisfy the conditional expressions of Equations (2) and (3), the processing device 104 uses the highest temperature Pmax in the high temperature region HTZ as the heat source position P source .

[0046] In this way, the thermal image positioning process 20 of the present invention can avoid the interference of non-human objects or non-the same human body when determining the temperature of the human body in the thermal image, so as to accurately locate the position of the forehead of the human body.

[0047] In addition, the above operation process of the thermal image positioning system 10 can be further summarized into a thermal image positioning process 70. In this way, when the thermal image positioning system 10 obtains the temperature array of the thermal image, it can obtain the forehead position and the corresponding temperature of the human body in the thermal image according to the thermal image positioning process 70.

[0048] The thermal image positioning process 70 includes the following steps:

[0049] Step 702: Start.

[0050] Step 704: Filter out the high-temperature values in the temperature array to generate a temperature array M.

[0051] Step 706: Use the N R *N R array mask to identify a high-temperature zone HTZ in the temperature array M.

[0052] Step 708: Obtain the center point Pcenter, the highest temperature point Pmax within the zone, and the average temperature Tavg within the zone.

[0053] Step 710: Use the N R *N R array mask to identify a region of interest ROI in the temperature array M.

[0054] Step 712: Determine whether there is a region of interest ROI. If so, execute Step 714; if not, execute Step 720.

[0055] Step 714: Obtain the center reference point P ROI of the region of interest and the corner points P1, P2.

[0056] Step 716: Determine whether the center reference point P ROI of the region of interest satisfies equation (1), (2), or (3). If so, execute Step 718; if not, execute Step 720.

[0057] Step 718: Determine the X-axis coordinate values of the corner points P1, P2 to determine the heat source position P source .

[0058] Step 720: Use the highest temperature Pmax within the high-temperature zone HTZ as the heat source position P source .

[0059] Regarding the operation process of the thermal image positioning process 70, please refer to the embodiments of the above thermal image positioning system 10 and thermal image positioning process 20, which will not be elaborated here.

[0060] It should be noted that the above embodiments describe the concept of the present invention, and those skilled in the art can make appropriate modifications accordingly and are not limited thereto. For example, the N T *N T array mask used to filter out the high-temperature values in the temperature array, and the N R *N RThe size of the array mask or the array mask used to determine the closest region, the size ratio of the pattern, the method and number of determining the corner points, the range of the center point interval Icenter, etc. can all be adjusted according to the settings of the user or the computer system. The up, down, left, and right position relationships are also directions appropriately set for easy understanding and are not limited to the above embodiments.

[0061] In summary, the embodiments of the present invention provide a thermal image positioning method and system. By corresponding to the temperature array of the thermal image, the head region and the position of the highest temperature point of the human body are calculated to accurately locate the forehead position.

[0062] The above are only the preferred embodiments of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope covered by the present invention.

[0063]

Symbol Explanation

[0064] 10: Thermal image positioning system

[0065] 102: Thermal imaging device

[0066] 104: Processing device

[0067] 1022: Sensor

[0068] 20, 70: Thermal image positioning process

[0069] 202~214, 702~720: Steps

[0070] A, B, A1_1~A1_9, A2_1~A2_9: Array elements

[0071] A1~A4: Closest region

[0072] B_1, B_2: Human body

[0073] FA_A, FA_B: Region fragmentation temperature array

[0074] H: Side length

[0075] Hleft: Left side length

[0076] Hright: Right side length

[0077] HTZ: High temperature region

[0078] Icenter: Center point interval

[0079] M: Temperature array

[0080] O: Origin

[0081] P1~P4: Corner points

[0082] Pcenter: Center point

[0083] Pmax: The highest temperature point within the area

[0084] P ROI : Reference point for the center of the region of interest

[0085] P source : Heat source location

[0086] ROI: Region of interest

[0087] T: Temperature threshold

[0088] Tavg: Average temperature within the area

[0089] Wtop: Upper side length

Claims

1. A thermal image positioning method for determining the position of a heat source in a thermal image, comprising: Obtaining a temperature array corresponding to the thermal image; Determining a region of interest (ROI) in the temperature array; Determining a center reference point of the region of interest; Determining a plurality of corner points corresponding to the center reference point of the region of interest; and Determining the position of the heat source based on at least one of the plurality of corner points.

2. The thermal image positioning method according to claim 1, further comprising: Filtering out a high temperature value to generate the temperature array; Judging a high temperature region in the temperature array; Determining a center point, a highest temperature point within the region, and an average temperature within the region of the high temperature region; and Determining the region of interest in the high temperature region by masking with a first array according to the center point, the highest temperature point within the region, and the average temperature within the region of the high temperature region.

3. The thermal image positioning method according to claim 2, wherein the step of filtering out the high temperature value to generate the temperature array comprises: Finding an array element higher than a temperature threshold, and masking and judging the array element and its adjacent array elements with a second array; and Filtering out the value of the array element and the values of its adjacent array elements, Among them, The temperature threshold is higher than an average human body temperature.

4. The thermal image positioning method according to claim 2, further comprising: Judging a closest region respectively from the center point of the high temperature region to the four corners of the first array mask according to a third array mask; and Respectively determining the corner points of the closest regions corresponding to the four corners of the first array mask.

5. The thermal image positioning method according to claim 4, wherein the first array mask is larger than the third array mask.

6. The thermal image positioning method according to claim 4, wherein the step of determining the position of the heat source based on at least one of the plurality of corner points comprises: Judging that the center reference point of the region of interest falls within a center point interval, and determining the position of the heat source based on at least one of the corner points of the first array mask, Among them, The center point interval is generated according to the highest temperature point within the region of the high temperature region and the first array mask.

7. The thermal image positioning method according to claim 6, wherein the step of determining the position of the heat source based on at least one of the plurality of corner points comprises: Based on a first corner point and a second corner point among the corner points corresponding to the four corners of the first array mask, respectively judging that the first corner point and the second corner point fall within the center point interval to determine the position of the heat source.

8. The thermal image positioning method according to claim 7, wherein when it is judged that both the first corner point and the second corner point fall outside the center point interval, the highest temperature within the high temperature region is used as the position of the heat source.

9. The thermal image positioning method according to claim 5, wherein the center reference point of the region of interest is determined according to an upper side length and a side length of a quadrilateral formed by the corner points corresponding to the four corners of the first array mask.

10. A thermal image positioning system for determining the position of a heat source in a thermal image, comprising: A thermal imaging device for generating a temperature array corresponding to the thermal image; And A processing device coupled to the thermal imaging device for determining a region of interest in the temperature array; determining a center reference point of the region of interest; determining a plurality of corner points corresponding to the center reference point of the region of interest; and determining the position of the heat source based on at least one of the plurality of corner points.

11. The thermal image positioning system according to claim 10, wherein the processing device is configured to filter out a high temperature value to generate the temperature array; determine a high temperature region in the temperature array; determine a center point, a highest temperature point within the region, and an average temperature within the region of the high temperature region; and determine the region of interest in the high temperature region by masking the high temperature region with a first array based on the center point, the highest temperature point within the region, and the average temperature within the region of the high temperature region.

12. The thermal image positioning system according to claim 11, wherein the processing device is configured to find array elements higher than a temperature threshold and mask the array elements and adjacent array elements of the array elements with a second array; and filter out the values of the array elements and the values of the adjacent array elements of the array elements, wherein the temperature threshold is higher than an average human body temperature.

13. The thermal image positioning system according to claim 11, wherein the processing device, based on a third array mask, determines a closest region from the center point of the high temperature region to each of the four corners of the first array mask; and determines the corner points of the closest regions corresponding to the four corners of the first array mask respectively.

14. The thermal image positioning system according to claim 13, wherein the first array mask is larger than the third array mask.

15. The thermal image positioning system according to claim 13, wherein the processing device is configured to determine that the center reference point of the region of interest falls within a center point interval, and determine the position of the heat source based on at least one of the corner points of the first array mask, wherein the center point interval is generated based on the highest temperature point within the high temperature region and the first array mask.

16. The thermal image positioning system according to claim 15, wherein the processing device is configured to determine the position of the heat source by respectively determining that a first corner point and a second corner point corresponding to the four corners of the first array mask fall within the center point interval.

17. The thermal image positioning system according to claim 16, wherein the processing device is configured to determine that both the first corner point and the second corner point fall outside the center point interval, and use the highest temperature within the high temperature region as the position of the heat source.

18. The thermal image positioning system according to claim 14, wherein the center reference point of the region of interest is determined based on an upper side length and a side length of a quadrilateral formed by the corner points corresponding to the four corners of the first array mask.

19. The thermal image positioning system according to claim 10, wherein the thermal image device includes a sensor for sensing a temperature, and the thermal imaging device generates the thermal image according to the temperature sensed by the sensor.

Images