Face temperature measurement image deduplication method, storage medium and computer equipment

Abstract

The invention provides a face temperature measurement image deduplication method, a storage medium and computer equipment, and relates to the technical field of face temperature measurement images. And rejecting the detected visible light image of which the width of the face frame is smaller than a first threshold value or the height is smaller than a second threshold value and the corresponding infrared image, rejecting the infrared image of which the pixel number is smaller than a third threshold value and the corresponding visible light image, and rejecting the infrared image of which the temperature is smaller than a fourth threshold value and the corresponding visible light image. If the similarity between the detected face and the face feature of the face data in the comparison database is smaller than a corresponding threshold value, the visible light image and the corresponding infrared image are stored, and if the similarity is larger than or equal to the corresponding threshold value, the visible light image and the corresponding infrared image are deleted. The method is used for reducing the memory pressure of a computer under the condition that a large number of repeated faces are generated in temperature detection, and can also reduce the workload of workers for counting alarm data.

Description

Technical Field

[0001] The present invention relates to the technical field of face temperature measurement image processing, and in particular to a face temperature measurement image deduplication method, a storage medium and a computer device. Background Art

[0002] Infrared temperature measurement technology is widely used in some public places, such as some ports. Through infrared imaging technology, pedestrians with abnormal temperatures can be found in real time, quickly and contactlessly, thereby effectively preventing the spread of epidemics. However, due to the characteristics of infrared images, faces are not very clear, and infrared-based face detection and comparison are extremely difficult. Visible light images have great advantages in face detection, recognition, and comparison. The combination of the two can quickly find pedestrians with abnormal temperatures.

[0003] However, when pedestrians pass through ports or harbors, the temperature is not instantaneous, but may last for a period of time. If an abnormal temperature occurs, an alarm will be triggered. At the same time, the background will keep saving images, which will cause the same person to save several or even many pictures, greatly increasing the redundancy of image data in the system and increasing the workload of statisticians. Summary of the invention

[0004] The purpose of the present invention is to provide a method for deduplicating face temperature measurement images, a storage medium and a computer device, which are intended to reduce the redundancy of image data in the system when a large number of duplicate faces are generated when temperature anomalies occur in uninterrupted temperature detection at all times, and reduce the temperature anomaly data that staff need to count. The specific technical solution is as follows:

[0005] A method for deduplication of a face temperature measurement image, the method comprising the following steps:

[0006] S100, acquiring a visible light image and a corresponding infrared image;

[0007] S200, performing face detection on the visible light image by using a deep learning method to obtain face frame coordinates and face key point coordinates, removing the visible light image and the corresponding infrared image whose width of the detected face frame is less than a first threshold or whose height is less than a second threshold, and retaining the visible light image and the corresponding infrared image whose width of the detected face frame is greater than or equal to the first threshold and whose height is greater than or equal to the second threshold;

[0008] S300, mapping the face frame of the retained visible light image to the corresponding infrared image, counting the number of pixels of the face in the infrared image, removing the infrared image and the corresponding visible light image whose number of pixels is less than a third threshold, and retaining the infrared image and the corresponding visible light image whose number of pixels is greater than or equal to the third threshold;

[0009] S400, performing infrared temperature measurement on the retained infrared images, eliminating the infrared images and corresponding visible light images whose temperatures are less than a fourth threshold, and retaining the infrared images and corresponding visible light images whose temperatures are greater than or equal to the fourth threshold;

[0010] S500, then calculate the Euclidean distance between the normalized key point coordinates of the detected face and the key point coordinates of the standard face, calculate the confidence between them based on the Euclidean distance, if the confidence is less than the fifth threshold, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the side face comparison database, if the similarity is less than the sixth threshold, add the face to the side face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the sixth threshold, only retain the first alarm visible light image and the corresponding infrared image; if the confidence is greater than or equal to the fifth threshold, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the front face comparison database, if the similarity is less than the seventh threshold, add the face to the front face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the seventh threshold, only retain the first alarm visible light image and the corresponding infrared image.

[0011] Furthermore, facial key points include left eye, right eye, nose, left corner of mouth and right corner of mouth.

[0012] Furthermore, the calculation formula for normalizing the coordinates of the detected facial key points is as follows:

[0013]

[0014] Among them, eye_x, eye1_y, eye2_x, eye2_y, nose_x, nose_y, mouth_x, mouth1_y, mouth2_x, mouth2_y represent the normalized horizontal and vertical coordinates of the left and right eyes, nose, and left and right corners of the mouth respectively, x1, y1 represent the horizontal and vertical coordinates of the left eye in the visible light image, x2, y2 represent the horizontal and vertical coordinates of the right eye in the visible light image, x3, y3 represent the horizontal and vertical coordinates of the nose in the visible light image, x4, y4 represent the horizontal and vertical coordinates of the left corner of the mouth in the visible light image, x5, y5 represent the horizontal and vertical coordinates of the right corner of the mouth in the visible light image, x, y represent the horizontal and vertical coordinates of the upper left corner of the face frame detected in the visible light image, and m, n represent the width and height of the detected face frame.

[0015] Furthermore, standard facial key points are obtained by pre-building a face dataset and then performing clustering and normalization processing.

[0016] Furthermore, the Euclidean distance calculation formula between the normalized key point coordinates of the detected face and the key point coordinates of the standard face is as follows:

[0017]

[0018] Among them, srceye1_x, srceye1_y represent the horizontal and vertical coordinates of the left eye after normalization of the standard face, srceye2_x, srceye2_y represent the horizontal and vertical coordinates of the right eye after normalization of the standard face, srcnose_x, srcnose_y represent the horizontal and vertical coordinates of the nose after normalization of the standard face, srcmouth1_x, srcmouth1_y represent the horizontal and vertical coordinates of the normalized left corner of the mouth of the standard face, srcmouth2_x, srcmouth2_y represent the horizontal and vertical coordinates of the normalized right corner of the mouth of the standard face, eye1_ω represents the Euclidean distance between the left eye of the detected face and the standard face, eye2-ω represents the Euclidean distance between the right eye of the detected face and the standard face, nose_ω represents the Euclidean distance between the nose of the detected face and the standard face, mouth1_ω represents the Euclidean distance between the left corner of the mouth of the detected face and the standard face, and mouth2_ω represents the Euclidean distance between the right corner of the mouth of the detected face and the standard face.

[0019] Furthermore, the calculation formula for the confidence of the detected face and the standard face is as follows:

[0020] λ=eye1_ω+eye2_ω+nose_ω+mouth1_ω+mouth2_ω

[0021]

[0022] Among them, α represents the confidence of the detected face and the standard face.

[0023] Furthermore, the calculation formula for the similarity between the detected face and the facial features of the faces in the profile or frontal face comparison database is as follows:

[0024] β(1)=f(1)*f j (1)

[0025] β(2)=f(2)*f j (2)

[0026] β(3)=f(3)*f j (3)

[0027]

[0028] β(t)=f(t)*f j (t)

[0029] η j =β(1)+β(2)+β(3)+...+β(t)

[0030] Among them, t represents the dimension of the extracted features, j represents the jth face in the profile or frontal face comparison database, f(t) represents the tth dimension feature of the detected face, and f j (t) represents the t-th dimension feature of the j-th face in the profile or frontal face comparison database, η j Represents the similarity between the detected face and the j-th face data in the profile or frontal face comparison database.

[0031] Furthermore, the side face or front face comparison database aligns the detected face or front face, extracts features, and saves them in a side face or front face table to facilitate subsequent comparison.

[0032] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the facial temperature measurement image deduplication method as described above are implemented.

[0033] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the facial temperature measurement image deduplication method as described above are implemented.

[0034] The present invention provides a method for deduplicating face temperature measurement images, which has the following beneficial effects:

[0035] The present invention provides a method for deduplicating a face temperature measurement image, which obtains a visible light image and a corresponding infrared image; performs face detection on the visible light image by a deep learning method to obtain face frame coordinates and face key point coordinates, removes the visible light image and the corresponding infrared image whose width of the detected face frame is less than a first threshold or whose height is less than a second threshold, and retains the visible light image and the corresponding infrared image whose width of the detected face frame is greater than or equal to the first threshold and whose height is greater than or equal to the second threshold; maps the face frame of the retained visible light image to the corresponding infrared image, counts the number of pixels of the face in the infrared image, removes the infrared image and the corresponding visible light image whose pixel number is less than a third threshold, and retains the infrared image and the corresponding visible light image whose pixel number is greater than or equal to the third threshold; performs infrared temperature measurement on the retained infrared image, removes the infrared image and the corresponding visible light image whose temperature is less than a fourth threshold, and retains the infrared image and the corresponding visible light image whose temperature is greater than or equal to the fourth threshold; then calculates the normalized face detection result. The Euclidean distance between the coordinates of the key points of the face and the coordinates of the key points of the standard face is used to calculate the confidence between them based on the Euclidean distance. If the confidence is less than the fifth threshold, face alignment is performed, face features are extracted, and the similarity between the detected face and the faces in the side face comparison database is calculated. If the similarity is less than the sixth threshold, the face is added to the side face comparison database and the visible light image and the corresponding infrared image are directly saved. If the similarity is greater than or equal to the sixth threshold, only the visible light image of the first alarm and the corresponding infrared image are retained; if the confidence is greater than or equal to the fifth threshold, face alignment is performed, face features are extracted, and the similarity between the detected face and the faces in the front face comparison database is calculated. If the similarity is less than the seventh threshold, the face is added to the front face comparison database and the visible light image and the corresponding infrared image are directly saved. If the similarity is greater than or equal to the seventh threshold, only the visible light image and the corresponding infrared image of the first alarm are retained, and the duplicates are no longer saved, thereby reducing the computer memory pressure and the workload of the staff in counting the alarm data. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a flow chart of a method for deduplication of facial temperature measurement images provided by the present invention;

[0037] Figure 2 is an overall flow chart of an embodiment of the present invention;

[0038] Figure 3 4 is a structural block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the accompanying drawings provided by the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. According to the following description, the advantages and features of the present invention will be more clear. It should be noted that the accompanying drawings are all in a very simplified form and are not in precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0040] Example 1

[0041] This embodiment provides a method for deduplicating face temperature measurement images. Figure 1 , 2 As shown, the method comprises the following steps:

[0042] S100: Acquire a visible light image and a corresponding infrared image.

[0043] Among them, the visible light image and the infrared image can be collected by a visible light camera and an infrared camera, which belongs to the existing technology and will not be described in detail here.

[0044] S200, performing face detection on the visible light image by a deep learning method to obtain face frame coordinates and face key point coordinates, discarding the visible light image and the corresponding infrared image whose width of the detected face frame is less than a first threshold R1 or whose height is less than a second threshold R2, and retaining the visible light image and the corresponding infrared image whose width of the detected face frame is greater than or equal to the first threshold R1 and whose height is greater than or equal to the second threshold R2.

[0045] Among them, face detection on the collected visible light images through deep learning methods belongs to the existing technology, which can be realized by the Yolo series model, MTCNN model, CNN model, etc., and will not be repeated here.

[0046] Specifically, the key points of the face include the left eye, the right eye, the nose, the left corner of the mouth, and the right corner of the mouth.

[0047] S300, mapping the face frame of the retained visible light image to the corresponding infrared image, counting the number of pixels of the face in the infrared image, eliminating the infrared image and the corresponding visible light image whose pixel numbers are less than a third threshold R3, and retaining the infrared image and the corresponding visible light image whose pixel numbers are greater than or equal to the third threshold R3.

[0048] S400, performing infrared temperature measurement on the retained infrared images, eliminating the infrared images and corresponding visible light images whose temperatures are less than a fourth threshold R4, and retaining the infrared images and corresponding visible light images whose temperatures are greater than or equal to the fourth threshold R4.

[0049] S500, then calculate the Euclidean distance between the normalized key point coordinates of the detected face and the key point coordinates of the standard face, calculate the confidence between them based on the Euclidean distance, if the confidence is less than the fifth threshold R5, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the side face comparison database, if the similarity is less than the sixth threshold R6, add the face to the side face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the sixth threshold R6, only retain the first alarm visible light image and the corresponding infrared image; if the confidence is greater than or equal to the fifth threshold R5, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the front face comparison database, if the similarity is less than the seventh threshold R7, add the face to the front face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the seventh threshold R7, only retain the first alarm visible light image and the corresponding infrared image.

[0050] Specifically, the calculation formula for normalizing the coordinates of the detected facial key points is as follows:

[0051]

[0052] Among them, eye1_x, eye1_y, eye2_x, eye2_y, nose_x, nose_y, mouth1_x, mouth1_y, mouth2_x, mouth2_y represent the normalized horizontal and vertical coordinates of the left and right eyes, nose, and left and right corners of the mouth respectively, x1, y1 represent the horizontal and vertical coordinates of the left eye in the visible light image, x2, y2 represent the horizontal and vertical coordinates of the right eye in the visible light image, x3, y3 represent the horizontal and vertical coordinates of the nose in the visible light image, x4, y4 represent the horizontal and vertical coordinates of the left corner of the mouth in the visible light image, x5, y5 represent the horizontal and vertical coordinates of the right corner of the mouth in the visible light image, x, y represent the horizontal and vertical coordinates of the upper left corner of the face frame detected in the visible light image, and m, n represent the width and height of the detected face frame.

[0053] Standard facial key points are obtained by pre-building a face dataset and then performing clustering and normalization processing.

[0054] The formula for calculating the Euclidean distance between the normalized key point coordinates of the detected face and the standard face key point coordinates is as follows:

[0055]

[0056] Among them, srceye1_x, srceye1_y represent the horizontal and vertical coordinates of the left eye after normalization of the standard face, srceye2_x, srceye2_y represent the horizontal and vertical coordinates of the right eye after normalization of the standard face, srcnose_x, srcnose_y represent the horizontal and vertical coordinates of the nose after normalization of the standard face, srcmouth1_x, srcmouth1_y represent the horizontal and vertical coordinates of the normalized left corner of the mouth of the standard face, srcmouth2_x, srcnouth2_y represent the horizontal and vertical coordinates of the normalized right corner of the mouth of the standard face, eye1_ω represents the Euclidean distance between the left eye of the detected face and the standard face, eye2_ω represents the Euclidean distance between the right eye of the detected face and the standard face, nose_ω represents the Euclidean distance between the nose of the detected face and the standard face, mouth1_ω represents the Euclidean distance between the left corner of the mouth of the detected face and the standard face, and mouth2_ω represents the Euclidean distance between the right corner of the mouth of the detected face and the standard face.

[0057] The calculation formula for the confidence of detected faces and standard faces is as follows:

[0058] λ=eye1_ω+eye2_ω+nose_ω+mouth1_ω+mouth2_ω

[0059]

[0060] Among them, α represents the confidence of the detected face and the standard face.

[0061] If the confidence level is less than the fifth threshold R5, it means that the detected face is a side face, and if the confidence level is greater than or equal to the fifth threshold R5, it means that the detected face is a front face.

[0062] The calculation formula for the similarity between the detected face and the facial features of the faces in the profile or frontal face comparison database is as follows:

[0063] β(1)=f(1)*f j (1)

[0064] β(2)=f(2)*f j (2)

[0065] β(3)=f(3)*f j (3)

[0066]

[0067] β(t)=f(t)*f j (t)

[0068] η j =β(1)+β(2)+β(3)+...+β(t)

[0069] Among them, t represents the dimension of the extracted features, j represents the jth face in the profile or frontal face comparison database, f(t) represents the tth dimension feature of the detected face, and f j (t) represents the t-th dimension feature of the j-th face in the profile or frontal face comparison database, η j Represents the similarity between the detected face and the j-th face data in the profile or frontal face comparison database.

[0070] The side face or front face comparison database aligns the detected face or front face, extracts features, and saves them in the side face or front face table to facilitate subsequent comparison.

[0071] The method for deduplicating face temperature measurement images provided by the present invention obtains a visible light image and a corresponding infrared image; performs face detection on the visible light image by a deep learning method to obtain face frame coordinates and face key point coordinates, removes the visible light image and the corresponding infrared image whose width of the detected face frame is less than a first threshold or whose height is less than a second threshold, and retains the visible light image and the corresponding infrared image whose width of the detected face frame is greater than or equal to the first threshold and whose height is greater than or equal to the second threshold; maps the face frame of the retained visible light image to the corresponding infrared image, counts the number of pixels of the face in the infrared image, removes the infrared image and the corresponding visible light image whose pixel number is less than a third threshold, and retains the infrared image and the corresponding visible light image whose pixel number is greater than or equal to the third threshold; performs infrared temperature measurement on the retained infrared image, removes the infrared image and the corresponding visible light image whose temperature is less than a fourth threshold, and retains the infrared image and the corresponding visible light image whose temperature is greater than or equal to the fourth threshold; then calculates the normalized face detection The Euclidean distance between the key point coordinates of the standard face and the key point coordinates of the standard face is calculated, and the confidence between them is calculated based on the Euclidean distance. If the confidence is less than the fifth threshold, face alignment is performed, face features are extracted, and the similarity between the detected face and the faces in the side face comparison database is calculated. If the similarity is less than the sixth threshold, the face is added to the side face comparison database and the visible light image and the corresponding infrared image are directly saved. If the similarity is greater than or equal to the sixth threshold, only the first alarm visible light image and the corresponding infrared image are retained; if the confidence is greater than or equal to the fifth threshold, face alignment is performed, face features are extracted, and the similarity between the detected face and the faces in the front face comparison database is calculated. If the similarity is less than the seventh threshold, the face is added to the front face comparison database and the visible light image and the corresponding infrared image are directly saved. If the similarity is greater than or equal to the seventh threshold, only the visible light image and the corresponding infrared image of the first alarm are retained, and the duplicates are no longer saved, thereby reducing the computer memory pressure and the workload of the staff in statistical alarm data.

[0072] Example 2

[0073] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for deduplication of facial temperature measurement images described above are implemented.

[0074] The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above-mentioned types of memory.

[0075] Example 3

[0076] This embodiment provides a computer device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the facial temperature measurement image deduplication method described above when executing the program.

[0077] like Figure 3 As shown, the computer device may include: at least one processor 71, such as a CPU (Central Processing Unit), at least one communication interface 73, a memory 74, and at least one communication bus 72. The communication bus 72 is used to realize the connection and communication between these components. The communication interface 73 may include a display screen (Display) and a keyboard (Keyboard), and the optional communication interface 73 may also include a standard wired interface and a wireless interface. The memory 74 may be a high-speed RAM memory (Random Access Memory, volatile random access memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 74 may optionally be at least one storage device located away from the aforementioned processor 71. The memory 74 stores application programs, and the processor 71 calls the program code stored in the memory 74 to execute any of the above method steps.

[0078] The communication bus 72 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 72 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0079] Among them, the memory 74 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 74 may also include a combination of the above types of memory.

[0080] The processor 71 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0081] The processor 71 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0082] Optionally, the memory 74 is also used to store program instructions. The processor 71 can call the program instructions to implement the face temperature measurement image deduplication method of the present invention.

[0083] Those skilled in the art should understand that the present invention can be implemented in many other specific forms without departing from the spirit and scope of the present invention. Based on the embodiments of the present invention, any changes and modifications made by ordinary technicians in the field of the present invention according to the above disclosure are within the scope of protection of the claims.

Claims

1. A method for deduplication of face temperature measurement images, characterized in that: The method comprises the following steps: S100, acquiring a visible light image and a corresponding infrared image; S200, performing face detection on the visible light image by using a deep learning method to obtain face frame coordinates and face key point coordinates, removing the visible light image and the corresponding infrared image whose width of the detected face frame is less than a first threshold or whose height is less than a second threshold, and retaining the visible light image and the corresponding infrared image whose width of the detected face frame is greater than or equal to the first threshold and whose height is greater than or equal to the second threshold; S300, mapping the face frame of the retained visible light image to the corresponding infrared image, counting the number of pixels of the face in the infrared image, removing the infrared image and the corresponding visible light image whose number of pixels is less than a third threshold, and retaining the infrared image and the corresponding visible light image whose number of pixels is greater than or equal to the third threshold; S400, performing infrared temperature measurement on the retained infrared images, eliminating the infrared images and corresponding visible light images whose temperatures are less than a fourth threshold, and retaining the infrared images and corresponding visible light images whose temperatures are greater than or equal to the fourth threshold; S500, then calculate the Euclidean distance between the normalized key point coordinates of the detected face and the key point coordinates of the standard face, calculate the confidence between them based on the Euclidean distance, if the confidence is less than the fifth threshold, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the side face comparison database, if the similarity is less than the sixth threshold, add the face to the side face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the sixth threshold, only retain the first alarm visible light image and the corresponding infrared image; if the confidence is greater than or equal to the fifth threshold, perform face alignment, extract facial features, calculate the similarity between the detected face and the faces in the front face comparison database, if the similarity is less than the seventh threshold, add the face to the front face comparison database and directly save the visible light image and the corresponding infrared image, if the similarity is greater than or equal to the seventh threshold, only retain the first alarm visible light image and the corresponding infrared image.

2. The method for deduplication of face temperature measurement images according to claim 1, characterized in that: The key points of the face include the left eye, right eye, nose, left corner of the mouth and right corner of the mouth.

3. The method for deduplication of face temperature measurement images according to claim 2, characterized in that: The calculation formula for normalizing the coordinates of the detected facial key points is as follows: Among them, eye1_x, eye1_y, eye2_x, eye2_y, nose_x, nose_y, mouth1_x, mouth1_y, mouth2_x, mouth2_y represent the normalized horizontal and vertical coordinates of the left and right eyes, nose, and left and right corners of the mouth respectively, x1, y1 represent the horizontal and vertical coordinates of the left eye in the visible light image, x2, y2 represent the horizontal and vertical coordinates of the right eye in the visible light image, x3, y3 represent the horizontal and vertical coordinates of the nose in the visible light image, x4, y4 represent the horizontal and vertical coordinates of the left corner of the mouth in the visible light image, x5, y5 represent the horizontal and vertical coordinates of the right corner of the mouth in the visible light image, x, y represent the horizontal and vertical coordinates of the upper left corner of the face frame detected in the visible light image, and m, n represent the width and height of the detected face frame.

4. The method for deduplication of face temperature measurement images according to claim 3, characterized in that: Standard facial key points are obtained by pre-building a face dataset and then performing clustering and normalization processing.

5. The method for deduplication of face temperature measurement images according to claim 4, characterized in that: The formula for calculating the Euclidean distance between the normalized key point coordinates of the detected face and the standard face key point coordinates is as follows: Among them, srceye1_x, srceye1_y represent the horizontal and vertical coordinates of the left eye after normalization of the standard face, srceye2_x, srceye2_y represent the horizontal and vertical coordinates of the right eye after normalization of the standard face, srcnose_x, srcnose_y represent the horizontal and vertical coordinates of the nose after normalization of the standard face, srcmouth1_x, srcmouth1_y represent the horizontal and vertical coordinates of the normalized left corner of the mouth of the standard face, srcmouth2_x, srcmouth2_y represent the horizontal and vertical coordinates of the normalized right corner of the mouth of the standard face, eye1_ω represents the Euclidean distance between the left eye of the detected face and the standard face, eye2_ω represents the Euclidean distance between the right eye of the detected face and the standard face, nose_ω represents the Euclidean distance between the nose of the detected face and the standard face, mouth1_ω represents the Euclidean distance between the left corner of the mouth of the detected face and the standard face, and mouth2_ω represents the Euclidean distance between the right corner of the mouth of the detected face and the standard face.

6. The method for deduplication of face temperature measurement images according to claim 5, characterized in that: The calculation formula for the confidence of detected faces and standard faces is as follows: λ=eye1_ω+eye2_ω+nose_ω+mouth1_ω+mouth2_ω Among them, α represents the confidence of the detected face and the standard face.

7. The method for deduplication of face temperature measurement images according to claim 6, characterized in that: The calculation formula for the similarity between the detected face and the facial features of the faces in the profile or frontal face comparison database is as follows: Among them, t represents the dimension of the extracted features, j represents the jth face in the profile or frontal face comparison database, f(t) represents the tth dimension feature of the detected face, and f j (t) represents the t-th dimension feature of the j-th face in the profile or frontal face comparison database, η j Represents the similarity between the detected face and the j-th face data in the profile or frontal face comparison database.

8. The method for deduplication of face temperature measurement images according to claim 7, characterized in that: The side face or front face comparison database aligns the detected face or front face, extracts features, and saves them in the side face or front face table to facilitate subsequent comparison.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for deduplication of facial temperature measurement images as described in any one of claims 1 to 8 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for deduplication of facial temperature measurement images as described in any one of claims 1 to 8 are implemented.

Images