WO2022083130A1 - Temperature measurement method and apparatus, electronic device, and storage medium - Google Patents

Abstract

The embodiments of the present disclosure provide a temperature measurement method and apparatus, an electronic device, and a storage medium. Said method comprises: acquiring an image to be processed, a temperature thermogram, and a homography matrix between the temperature thermogram and said image; performing face detection on said image to obtain a first face region; on the basis of the homography matrix, determining, from the temperature thermogram, a pixel point region corresponding to the first face region, so as to obtain a second face region; and on the basis of the temperatures of pixel points contained in the second face region, determining the temperature of a temperature measurement object corresponding to the first face region.

Description

Temperature measurement method and device, electronic equipment and storage medium

CROSS-REFERENCE TO RELATED APPLICATIONS

This disclosure is based on the Chinese patent application with the application number of 202011148359.5 and the application date of October 23, 2020, and the application name is "temperature measurement method and device, electronic equipment and storage medium", and claims the priority of the Chinese patent application, The entire contents of this Chinese patent application are hereby incorporated by reference into the present disclosure.

technical field

The embodiments of the present disclosure relate to the technical field of temperature measurement, and in particular, to a temperature measurement method and device, an electronic device, and a storage medium.

Background technique

At present, the demand for non-contact temperature measurement in the field of temperature measurement has exploded. Non-contact temperature measurement can effectively avoid cross-infection of the crowd and play a positive role in epidemic prevention.

In the current non-contact temperature measurement method, the temperature measurement device obtains the face area in the temperature heat map by performing face detection on the collected temperature heat map. Based on the temperature of the pixels included in the face area in the temperature heat map, the temperature of the temperature measurement object can be obtained. However, the accuracy of the temperature of the temperature measurement object obtained by this method is relatively low.

SUMMARY OF THE INVENTION

Embodiments of the present disclosure provide a temperature measurement method and device, an electronic device, and a storage medium.

An embodiment of the present disclosure provides a temperature measurement method, the method includes: acquiring a to-be-image, a temperature heat map, and a homography matrix between the temperature heat map and the to-be-image; face detection to obtain a first face area; determine a pixel area corresponding to the first face area from the temperature heat map based on the homography matrix to obtain a second face area; based on the first face area The temperature of the pixels included in the two face regions determines the temperature of the temperature measurement object corresponding to the first face region.

In some possible implementation manners, the determining a pixel area corresponding to the first face area from the temperature heat map based on the homography matrix, to obtain a second face area, includes: based on the The homography matrix and the four corners of the face frame containing the first face region, and four pixels corresponding to the four corners are determined from the temperature heat map; based on the The quadrilateral area determined by the four pixel points determines the second face area.

In some possible implementations, performing face detection on the to-be-processed image to obtain a first face region includes: performing face detection on the to-be-processed image to obtain at least one first face frame; The pixel area included in the second face frame in the at least one first face frame is used as the first face area; the resolution of the pixel area included in the second face frame is greater than the resolution threshold face frame.

In some possible implementations, the number of the first face frames is greater than 1, and the at least one first face frame includes a third face frame and a fourth face frame; the at least one first face frame The pixel area included in the second face frame in the face frame is used as the first face area, including: the overlap ratio between the third face frame and the fourth face frame is less than or When it is equal to the overlap rate threshold, the pixel point area included in any one of the third face frame and the fourth face frame is used as the first face area.

In some possible implementation manners, the pixel point area included in any one of the face frames in the third face frame and the fourth face frame as the first face area includes: The pixel area contained in the face frame with the largest size measure in the third face frame and the fourth face frame is used as the first face area; the size measure of the third face frame is The maximum value of the side length of the third human face frame; the size measurement of the fourth human face frame is the maximum value of the side length of the fourth human face frame.

In some possible implementation manners, the determining the temperature of the temperature measurement object corresponding to the first human face region based on the temperature of the pixels included in the second human face region includes: measuring the temperature of the first human face region. Performing forehead detection on the face area to obtain the forehead detection result of the first face area; in the case where the forehead detection result of the first face area is that the forehead area in the first face area is not blocked , the temperature of the temperature measurement object is obtained based on the temperature of the pixels included in the third face area; the third face area is the area corresponding to the forehead area in the second face area.

In some possible implementation manners, the determining the second face area based on the quadrilateral area determined by the four pixel points includes: acquiring an intersection point of diagonal lines of the quadrilateral area; The distance from the intersection point is taken as the first distance; the first point is the pixel point closest to the intersection point among the four pixel points; taking the intersection point as the center of the circle and the first distance as the radius, constructing a first area; determining the intersection of the first area and the quadrilateral area to obtain a second area; selecting an area including the intersection point from the second area as the second face area.

In some possible implementation manners, the determining the second face area based on the quadrilateral area determined by the four pixel points includes: acquiring an intersection of diagonal lines of the quadrilateral area; determining the quadrilateral area The largest inscribed area of the area; the largest inscribed area is a rectangular area or a circular area including the intersection point; the largest inscribed area is used as the second face area.

In some possible implementation manners, the determining the largest inscribed area of the quadrilateral area includes: selecting the second largest abscissa from the abscissas of the four pixel points to obtain the first abscissa, and obtaining the first abscissa from the From the abscissas of the four pixels, the third largest abscissa is selected to obtain the second abscissa, the second largest ordinate is selected from the ordinates of the four pixels to obtain the first ordinate, and the first ordinate is obtained from the four ordinates. Among the ordinates of the pixel point, the third largest ordinate is selected to obtain the second ordinate; the second point is determined based on the first abscissa and the first ordinate, and the second point is determined based on the first abscissa and the second The ordinate determines the third point, the fourth point is determined based on the second abscissa and the first ordinate, and the fifth point is determined based on the second abscissa and the second ordinate; The area determined by the point, the third point, the fourth point and the fifth point is used as the maximum inscribed area.

In some possible implementation manners, the embodiments of the present disclosure further provide a temperature measurement device, the device includes: an acquisition part configured to acquire an image to be processed, a temperature heat map, and the temperature heat map and the to-be-processed heat map processing the homography matrix between the images; the detection part is configured to perform face detection on the to-be-processed image to obtain a first face region; the first processing part is configured to obtain a first face region based on the homography matrix; A pixel area corresponding to the first face area is determined in the temperature heat map to obtain a second face area; a second processing part is configured to be based on the temperature of the pixels included in the second face frame , and determine the temperature of the temperature measurement object corresponding to the first face region.

In some possible implementation manners, the first processing part is further configured to: based on the homography matrix and four corner points of a face frame including the first face region, extract the temperature from the temperature Four pixel points corresponding to the four corner points are determined in the heat map; based on the quadrilateral area determined by the four pixel points, the second face area is determined.

In some possible implementation manners, the detection part is further configured to: perform face detection on the to-be-processed image to obtain at least one first face frame; The pixel area included in the second face frame is used as the first face area; the resolution of the pixel area included in the second face frame is a face frame whose resolution is greater than the resolution threshold.

In some possible implementations, the number of the first face frames is greater than 1, and the at least one first face frame includes a third face frame and a fourth face frame; the detection part is further configured is: when the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, the third face frame and the fourth face frame are divided into Any pixel area included in the face frame is used as the first face area.

In some possible implementation manners, the detection part is further configured to: use the pixel area included in the face frame with the largest size measurement in the third face frame and the fourth face frame as the the first face area; the size measurement of the third face frame is the maximum value of the side length of the third face frame; the size measurement of the fourth face frame is the fourth face frame the maximum side length of .

In some possible implementation manners, the second processing part is further configured to: perform forehead detection on the first face region to obtain a forehead detection result of the first face region; The forehead detection result of the face area is that the forehead area in the first face area is not blocked, and the temperature of the temperature measurement object is obtained based on the temperature of the pixels included in the third face area; The third face area is an area corresponding to the forehead area in the second face area.

In some possible implementation manners, the first processing part is further configured to: obtain the intersection of the diagonal lines of the quadrilateral area; take the distance between the first point and the intersection as the first distance; The first point is the pixel point closest to the intersection point among the four pixel points; taking the intersection point as the center of the circle and the first distance as the radius, construct a first area; determine the first area and the The intersection of the quadrilateral regions is obtained to obtain a second region; the region including the intersection point is selected from the second region as the second face region.

In some possible implementation manners, the first processing part is further configured to: obtain the intersection of the diagonal lines of the quadrilateral area; determine the maximum inscribed area of the quadrilateral area; the maximum inscribed area is A rectangular area or a circular area including the intersection point; the largest inscribed area is taken as the second face area.

In some possible implementation manners, the first processing part is further configured to: select the second largest abscissa from the abscissas of the four pixel points to obtain the first abscissa, and obtain the first abscissa from the abscissas of the four pixels. From the abscissa of the point, select the third largest abscissa to obtain the second abscissa, select the second largest ordinate from the ordinate of the four pixel points to obtain the first ordinate, and obtain the first ordinate from the ordinate of the four pixel points. Selecting the third largest ordinate among the ordinates to obtain the second ordinate; determining the second point based on the first abscissa and the first ordinate, and determining the second point based on the first abscissa and the second ordinate For the third point, a fourth point is determined based on the second abscissa and the first ordinate, and a fifth point is determined based on the second abscissa and the second ordinate; The area determined by the third point, the fourth point and the fifth point is used as the maximum inscribed area.

In some possible implementation manners, the embodiments of the present disclosure further provide a processor configured to execute the method according to the above-mentioned first aspect and any one of its possible implementation manners.

In some possible implementations, embodiments of the present disclosure also provide an electronic device, including: a processor, a sending device, an input device, an output device, and a memory, the memory is configured to store computer program codes, the computer The program code includes computer instructions, and when the processor executes the computer instructions, the electronic device executes the method according to the first aspect and any one of possible implementations thereof.

In some possible implementation manners, embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and the computer program includes program instructions, where the program instructions are In the case of execution by a processor, the processor is caused to execute the method according to the first aspect and any one of possible implementation manners thereof.

In some possible implementation manners, the embodiments of the present disclosure also provide a computer program product, the computer program product includes a computer program or instructions, and when the computer program or instructions are run on a computer, the The computer performs the method of the first aspect and any of its possible implementations.

It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not limiting of the disclosed embodiments.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the background technology, the accompanying drawings required in the embodiments or the background technology of the present disclosure will be described below.

The accompanying drawings, which are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description, serve to explain the technical solutions of the embodiments of the present disclosure.

1 is a schematic flowchart of a temperature measurement method provided by an embodiment of the present disclosure;

FIG. 2 is a schematic structural diagram of a temperature measuring device according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a hardware structure of a temperature measuring device according to an embodiment of the present disclosure.

Detailed ways

In order to make those skilled in the art better understand the solutions of the embodiments of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described The embodiments are only some of the embodiments of the present disclosure, but not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the embodiments of the present disclosure.

The terms "first", "second" and the like in the description and claims of the embodiments of the present disclosure and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. Furthermore, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or parts is not limited to the listed steps or parts, but optionally also includes unlisted steps or parts, or optionally also includes For other steps or parts inherent to these processes, methods, products or devices.

Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present disclosure. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor a separate or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

At present, the demand for non-contact temperature measurement in the field of temperature measurement has exploded. The non-contact temperature measurement algorithm can effectively avoid the cross-infection of the crowd and play a positive role in epidemic prevention.

The commonly used temperature measurement method is to directly use the temperature heat map generated by the temperature measurement module of the non-contact temperature measurement device to perform face detection, and obtain the face area in the temperature heat map. Then, according to the temperature of the pixels contained in the face area in the temperature heat map, the temperature of the temperature measurement object is obtained. The temperature heat map of the temperature measurement modules produced by each manufacturer is quite different, and it is very expensive to make a general heat map face detection model. Moreover, the face detection models of non-contact temperature measurement devices are not trained based on temperature heatmaps. The temperature heat map carries less information, and the effect of face detection will be poor, resulting in inaccurate face areas. Therefore, according to the temperature of the pixel area included in the inaccurate face area in the temperature heat map, the temperature of the temperature measurement object obtained is also inaccurate. This temperature measurement method not only requires a large labor cost, but also has low detection efficiency, which can easily lead to inaccurate temperature measurement results.

Based on this, an embodiment of the present disclosure provides a temperature measurement method implemented by a temperature measurement device. The execution body of the embodiment of the present disclosure is a temperature measuring device. Optionally, the temperature measuring device may be one of the following: a mobile phone, a computer, a tablet computer, and an access control device. The embodiments of the present disclosure will be described below with reference to the accompanying drawings in the embodiments of the present disclosure.

Please refer to FIG. 1 , which is a schematic flowchart of a temperature measurement method provided by an embodiment of the present disclosure.

Step 101: Acquire an image to be processed, a temperature heat map, and a homography matrix between the above temperature heat map and the above image to be processed.

In some possible implementations, the image to be processed may be any image. For example, the image to be processed can contain a human face or a human face and an object. The embodiments of the present disclosure do not limit the content included in the image to be processed.

The temperature heat map is a pseudo-color map corresponding to the image to be processed that carries temperature information. Each pixel of the temperature heat map carries corresponding temperature information. The darker the color of the pixels in the temperature heat map, the higher the temperature. That is to say, according to the temperature of the pixel points in the temperature heat map, the temperature of the temperature measurement object in the image to be processed can be obtained.

In a possible implementation manner of acquiring the image to be processed, the temperature measuring device is equipped with an RGB camera. The temperature measuring device acquires the image to be processed through the RGB camera acquisition. Optionally, the temperature measuring device is assembled with the access control device. The temperature measurement devices are artificial intelligence (AI) infrared imagers, non-contact temperature measurement products such as security gates (such products are mainly placed in stations, airports, subways, shops, supermarkets, schools, company halls and community gates). these crowded scenes).

In another possible implementation manner of acquiring the image to be processed, the temperature measuring device is equipped with a monitoring camera. The video stream sent by the surveillance camera is decoded, and the RGB image obtained by the decoding process is used as the image to be processed.

In one possible implementation manner of acquiring the temperature thermal map, the temperature measuring device is equipped with an infrared imaging device. The temperature measuring device acquires a temperature thermal map through the infrared imaging camera of the imaging device.

In some possible implementations, a homography transformation is equivalent to a projective transformation, changing the position and shape of an object. The homography matrix (Homograph, H) is a pre-calibrated homography mapping transformation parameter. The embodiment of the present disclosure adopts a traditional method to obtain the homography matrix between the temperature heat map and the image to be processed. The traditional method to obtain the homography matrix needs to go through the following steps: extract the feature points of each image; extract the descriptor corresponding to each feature point; find the matching feature point pair in the two images by matching the feature point descriptor (there may be Error matching); use the RANSAC (Random Sample Consensus, random sampling consistent) algorithm to eliminate false matching; solve the equation system, and calculate the homography matrix. The temperature heat map is an image captured by an infrared imaging device, and the image to be processed is an image captured by an RGB imaging device.

In an implementation manner that may acquire the homography matrix between the temperature heat map and the image to be processed, an RGB camera and an infrared imaging camera are installed on the temperature measuring device. Place a square card about 1 meter in front of the temperature measuring device, and take two images first through the RGB camera and the infrared imaging camera. Then, get the RGB image of the square card through the RGB camera. An infrared imaging image of the square card is obtained by an infrared imaging camera. By extracting the local feature points of the RGB image of the square card and the infrared imaging image of the square card, the algorithms that can be used are: scale-invariant feature transform (SIFT), speed up roust feature, SURF), corner detection (Features fromaccelerated segment test, FAST), etc., which are not limited in the embodiments of the present disclosure. Then, extract the descriptor corresponding to each local feature point, and find the matching feature point pair in the RGB image of the square card and the infrared imaging image of the square card by matching the feature point descriptor. Because there may be wrong matches, the RANSAC algorithm is used to eliminate the wrong matches, solve the corresponding equation system, and calculate the homography matrix (Homograph). It should be understood that the positions and angles of the RGB cameras and infrared imaging cameras of each temperature measurement device may be different, and the homography matrices obtained by the RGB cameras and infrared imaging cameras of different temperature measurement devices are different. Therefore, each temperature measuring device needs to be calibrated by a square card. After obtaining the image captured by the RGB camera and the image captured by the infrared imaging camera, the homography matrix is calculated by the above method.

Step 102, performing face detection on the above-mentioned to-be-processed image to obtain a first face region;

In some possible implementation manners, face detection is used to determine whether the above-mentioned image to be processed contains a face region. By performing face detection on the above-mentioned image to be processed, the temperature measuring device can determine whether the above-mentioned image to be processed includes a face region.

The temperature measurement device determines that the image to be processed includes a face region representation, and the image to be processed includes a temperature measurement object. In the case that the above image to be processed includes a face region, the first face region is obtained. It should be understood that the first face area is the face area corresponding to the temperature measurement object in the above image to be processed.

In a possible implementation manner of face detection, the temperature measurement device obtains face feature data of the to-be-processed image by performing facial feature extraction processing on the to-be-processed image. Further, it can be determined based on the face feature data whether the image to be processed includes a face region.

In another possible implementation manner of face detection, the temperature measurement device uses a face detection neural network to process the image to be processed, so as to determine whether the image to be processed contains a face region. The face detection neural network is obtained by training a first training image with label information as training data, wherein the label information includes whether the first training image contains a human face.

In another possible implementation manner of face detection, feature extraction is performed on the image to be processed through a pre-trained neural network to obtain feature data, and the pre-trained neural network identifies whether the image to be processed Contains human faces. By performing feature extraction processing on the image to be processed, in the case where it is determined that the image to be processed contains a human face in the feature extraction data, the position of the face frame of the image to be processed is determined, that is, the detection of the face is realized, and the image to be processed is detected. Face detection can be achieved through convolutional neural networks. By using multiple images with labeled information as training data, the convolutional neural network is trained, so that the trained convolutional neural network can complete the face detection on the image. The annotation information of the images in the training data is the face and the position of the face. In the process of using the training data to train the convolutional neural network, the convolutional neural network extracts the feature data of the image from the image, and determines whether there is a face in the image based on the feature data. The feature data of the image gets the position of the face. The annotation information is used as the supervision information to supervise the results obtained by the convolutional neural network in the training process, and the parameters of the convolutional neural network are updated to complete the training of the convolutional neural network. In this way, the image to be processed can be processed using the trained convolutional neural network to obtain the position of the face in the image to be processed.

In yet another possible implementation manner of face detection, face detection may be implemented through a face detection algorithm, wherein the face detection algorithm may be one of the following: face detection based on coarse histogram segmentation and singular value features algorithm, face detection based on binary wavelet transform, neural network method based on probability decision (probabilistic decision based neural network, PDBNN), hidden Markov model method (Hidden Markov Model, HMM), etc. The face detection algorithm for implementing face detection is not limited.

Step 103: Determine the pixel area corresponding to the above-mentioned first face area from the above-mentioned temperature heat map based on the above-mentioned homography matrix, and obtain the second face area;

The pixel points corresponding to the pixel points in the image to be processed can be determined from the temperature heat map through the homography matrix.

For example, assuming that the coordinates of pixel A in the image to be processed are (2, 4), the pixel corresponding to pixel A in the temperature heat map is pixel D, and the homography matrix

Convert the coordinates of pixel A into homogeneous coordinates (2, 4, 1), and the corresponding matrix form B is [2, 4, 1]. Then, multiply the homogeneous coordinates of the A pixel by the homography matrix H to get the matrix

Divide the first value 13 of matrix C by the third value 55, the abscissa of pixel D is 13/55; the second value 34 of matrix C is divided by the third value 55, and the abscissa of pixel D is obtained. The ordinate is 34/55. Therefore, the coordinates of the pixel point D corresponding to the pixel point A in the temperature heat map are (13/55, 34/55).

Based on the homography matrix, the image processing device can determine the pixel points corresponding to the pixel points on the contour of the first face area from the temperature heat map, and then determine the pixel point area corresponding to the first face area from the temperature heat map , which is the second face area.

Step 104: Obtain the temperature of the temperature measurement object corresponding to the first face region based on the temperature of the pixels included in the second face region.

In some possible implementations, the temperature measuring device can determine the temperature corresponding to any pixel point in the temperature thermogram. When the above-mentioned temperature heat map includes a second face area, the temperature measuring device can obtain the temperature of the temperature measurement object corresponding to the first face area based on the temperature of the pixel points included in the second face area.

In some possible implementation manners, the temperature measurement device uses the average temperature of at least one pixel point in the second face region as the temperature of the temperature measurement object. For example, the pixel points of the second face region include: pixel point a and pixel point b, wherein the temperature corresponding to pixel point a is 36.9 degrees, and the temperature corresponding to pixel point b is 36.3 degrees. The temperature measuring device can take the average value (36.6 degrees) of the temperature corresponding to pixel point a and the temperature corresponding to pixel point b as the temperature of the temperature measurement object; the temperature measuring device can also use the temperature corresponding to pixel point a (36.9 degrees) ) as the temperature of the temperature measurement object; the temperature measurement device can also use the temperature (36.3 degrees) corresponding to the pixel point b as the temperature of the temperature measurement object.

In another possible implementation manner, the temperature measurement device uses the highest temperature value corresponding to the pixel point in the second face region as the temperature of the temperature measurement object. For example, the pixels in the second face area include: pixel a, pixel b, and pixel c, where the temperature corresponding to pixel a is 36.9 degrees, the temperature corresponding to pixel b is 36.3 degrees, and the temperature corresponding to pixel a is 36.3 degrees. c corresponds to a temperature of 37 degrees. The temperature measurement device can take the temperature corresponding to the pixel point a, the temperature corresponding to the pixel point b, and the maximum value (37 degrees) of the temperature corresponding to the pixel point c as the temperature of the temperature measurement object.

In another possible implementation manner, the temperature measuring device first calculates the average temperature corresponding to the pixel points in the second face region, and selects at least one pixel point whose difference from the average temperature value is less than or equal to a threshold value pixel points, and the average value of the temperature corresponding to at least one pixel point is used as the temperature of the temperature measurement object. For example, the pixel points of the second face region include: pixel point a, pixel point b, pixel point c, and pixel point d. Among them, the temperature corresponding to pixel a is 36.9 degrees, the temperature corresponding to pixel b is 36.3 degrees, the temperature corresponding to pixel c is 37.9 degrees, and the temperature corresponding to pixel d is 35.2 degrees. The temperature measuring device calculates that the average temperature corresponding to the pixels in the second face area is 36.625 degrees. When the threshold is 1 degree, the difference between the temperature corresponding to pixel c and the average temperature (1.275 degrees) and the difference between the temperature corresponding to pixel d and the average temperature (1.375 degrees) are both If it is greater than 1 degree, the temperature measuring device selects at least one pixel point from the pixel point a and the pixel point b. Further, the temperature measuring device can use the average value (36.6 degrees) of the temperature corresponding to the pixel point a and the temperature corresponding to the pixel point b as the temperature of the temperature measurement object; the temperature measuring device can also use the temperature corresponding to the pixel point a. (36.9 degrees) as the temperature of the temperature measurement object; the temperature measurement device can also use the temperature (36.3 degrees) corresponding to the pixel point b as the temperature of the temperature measurement object.

In some possible implementations, the temperature measuring device can accurately determine the face region of the temperature measurement object from the to-be-processed image by performing face detection on the to-be-processed image to obtain the first face region. Based on the homography matrix between the image to be processed and the temperature heat map and the position of the first face region in the image to be processed, determining the face region of the temperature measurement object from the temperature heat map can improve the performance from the temperature heat map. Determine the accuracy of the face area of the temperature measurement object to obtain a second face area. In this way, the temperature measurement device obtains the temperature of the temperature measurement object based on the temperature of the second face region, which can improve the accuracy of the temperature measurement object.

As an optional implementation manner, in order to determine the pixel area corresponding to the first face region in the temperature heat map, the position of the first face region needs to be determined first. The position of the first face region is determined by the coordinates of the four corners of the face frame including the first face region. When the face detection obtains the first face area, it can also obtain the face frame including the first face area. Therefore, the temperature measurement device performs the following steps in the process of executing step 102:

Step 21, performing face detection on the above-mentioned to-be-processed image to obtain at least one first face frame;

In some possible implementation manners, when there are multiple faces in the image to be processed, by performing face detection on the image to be processed, according to the size of the face area in the above image to be processed, output a plurality of people including the face area face frame. That is, performing face detection on the above-mentioned image to be processed to obtain at least one first face frame. It should be understood that the size of the face frame is determined according to the size of the included face area. The shape of each face frame in the at least one first face frame is a rectangle, and the position of each face frame is determined by the coordinates of four corners of the face frame.

Step 22, taking the pixel area included in the second face frame in the above-mentioned at least one first face frame as the first face area; the resolution of the pixel area included in the second face frame is greater than the resolution threshold the face frame;

In some possible implementation manners, a face frame whose resolution of the pixel area included in the at least one first face frame is greater than a resolution threshold is selected to obtain the second face frame. In the embodiment of the present disclosure, the face pixels for face detection are at least 60×60 or more. That is, the resolution threshold is at least 60x60. For example, when the face pixels of the first face area are 56x56, the accuracy of obtaining the first face area cannot be determined after face detection, nor can it be determined in the temperature heat map that it is related to the first face area. The accuracy of the corresponding second face region. Therefore, the resolution threshold is at least set to be greater than or equal to 60×60, but the value of the resolution threshold is not limited in this embodiment of the present disclosure.

In a possible implementation, when the number of at least one first face frame is 1 and the resolution of the pixel area included in the face frame is greater than the resolution threshold, a second face frame is obtained, and the first face frame is The two face frames are used as face frames of the temperature measurement object in the above image to be processed.

As an optional implementation manner, by performing face detection on the image to be processed, there may be multiple face frames obtained. Because the embodiment of the present disclosure involves single-person temperature measurement, one needs to be selected from multiple face frames. When the number of the first face frames is greater than 1, the at least one first face frame includes a third face frame and a fourth face frame, and the temperature measurement device is performing The process of using the pixel area included in the two-face frame as the above-mentioned first face area includes the following steps:

Step 23, in the case that the overlap ratio between the above-mentioned third face frame and the above-mentioned fourth face frame is less than or equal to the overlap ratio threshold, place any one of the above-mentioned third face frame and the above-mentioned fourth face frame. The pixel area contained in the face frame is used as the above-mentioned first face area;

In some possible implementation manners, when the number of the above-mentioned first face frames is greater than 1, a face frame including the first face region is selected from at least one first face frame. In the above-mentioned at least one face frame of the first face frame in the above image to be processed, there is at least one face frame different from it, so that the pixel area contained in the two face frames has overlapping parts, The pixel point area included in any one of the at least one first face frame cannot be used as the above-mentioned first face area. That is to say, in any face frame of at least one first face frame in the above image to be processed, there is at least one face frame different from it, so that the overlap ratio between the two face frames is greater than the overlap ratio In the case of the threshold value, the pixel point area included in any one of the at least one first face frame cannot be used as the above-mentioned first face area.

In the case where the overlap ratio between any two face frames in at least one first face frame is less than or equal to the overlap ratio threshold, that is, there is a distance between each face frame in the at least one first face frame, and That is, the area of the overlapping part of the pixel area contained in any two face frames is 0, that is to say, the overlap rate between any two face frames is 0%. Then, the pixel area included in any one face frame in the at least one first face frame can be used as the above-mentioned first face area. Therefore, set the overlap rate threshold to 0.

In a possible implementation manner, the number of at least one first face frame is 2, and the two face frames are set as the third face frame and the fourth face frame. Calculate the overlap ratio between the third face frame and the fourth face frame, and in the case that the overlap ratio between the third face frame and the fourth face frame is determined to be less than or equal to the overlap ratio threshold The pixel point area contained in any one of the face frame and the fourth face frame is used as the above-mentioned first face area. For example, at least one first face frame has two face frames: face frame b and face frame c. Calculate the overlap rate between face frame b and face frame c. In the case where the overlap ratio between face frame b and face frame c is less than or equal to the overlap ratio threshold, that is, the pixel area contained in face frame b does not overlap with the pixel area contained in face frame c. . Therefore, the pixel area included in the face frame b can be used as the first face area, and the pixel area included in the face frame c can also be used as the first face area. In the case where the overlap ratio between the face frame b and the face frame c is greater than the overlap ratio threshold, that is, the pixel area included in the face frame b and the pixel area included in the face frame c overlap. Therefore, the pixel area included in the face frame b cannot be used as the first face area, and the pixel area included in the face frame c cannot be used as the first face area either.

In another possible implementation manner, the number of at least one first face frame is 3, and the three face frames are set as the third face frame, the fourth face frame and the fifth face frame. Calculate the overlap rate between the third face frame and the fourth face frame, the overlap ratio between the third face frame and the fifth face frame, and the overlap between the fourth face frame and the fifth face frame. overlap rate.

A possible situation is that the overlap ratio between the third face frame and the fourth face frame is greater than the overlap ratio threshold, and the overlap ratio between the third face frame and the fifth face frame is less than or equal to the overlap ratio threshold, The overlap ratio between the fourth face frame and the fifth face frame is less than or equal to the overlap ratio threshold. That is to say, the pixel area included in the third face frame and the pixel area included in the fourth face frame overlap, and the pixel area included in the fifth face frame and the pixels included in the third face frame overlap. There is no overlapping part of the area, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame cannot be used as the first face area, and the pixel area included in the fourth face frame cannot be used as the first face area, only the pixel area included in the fifth face frame. Can be used as the first face area.

In another possible situation, the overlap rate between the third face frame and the fourth face frame is greater than the overlap rate threshold, the overlap rate between the third face frame and the fifth face frame is greater than the overlap rate threshold, and the third face frame is greater than the overlap rate threshold. The overlap ratio between the fourth face frame and the fifth face frame is less than or equal to the overlap ratio threshold. That is to say, the pixel area included in the third face frame and the pixel area included in the fourth face frame overlap, and the pixel area included in the fifth face frame and the pixels included in the third face frame overlap. The area has an overlapping part, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame cannot be used as the first face area, the pixel area included in the fourth face frame cannot be used as the first face area, and the pixel area included in the fifth face frame cannot be used as the first face area. as the first face area.

In another possible situation, the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, and the overlap ratio between the third face frame and the fifth face frame is less than or equal to the overlap rate threshold, the overlap rate between the fourth face frame and the fifth face frame is less than or equal to the overlap rate threshold. That is to say, the pixel area contained in the third face frame does not overlap with the pixel area contained in the fourth face frame, and the pixel area contained in the fifth face frame and the pixels contained in the third face frame do not overlap. There is no overlapping part of the area, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame can be used as the first face area, the pixel area included in the fourth face frame can be used as the first face area, and the pixel area included in the fifth face frame can also be used as the first face area. as the first face area.

In another possible implementation manner, in the case where the resolution of the pixel region contained in the above at least one first face frame is greater than the number of face frames with a resolution threshold greater than 1, the resolution is greater than the resolution threshold. Select the face frame containing the first face area in at least one of the first face frames. The resolution of the pixel area included in at least one first face frame in the above-mentioned to-be-processed image is greater than the number of face frames with a resolution threshold greater than 1, and the resolution of the pixel area included in the at least one first face frame is greater than 1. For any face frame in the face frame larger than the resolution threshold, there is at least one face frame different from it, so that when the pixel areas contained in the two face frames overlap, at least one of the first face frames cannot be combined. The resolution of the pixel point region included in the face frame is greater than the resolution threshold, and the pixel point region included in any face frame in the face frame is used as the first face region. That is to say, in any face frame in which the resolution of the pixel area contained in the at least one first face frame is greater than the resolution threshold, there is at least one face frame different from it, so that these two When the overlap ratio between the face frames is greater than the overlap ratio threshold, the pixels included in any face frame in the face frame whose resolution of the pixel area included in at least one first face frame is greater than the resolution threshold cannot be used. The point area is used as the above-mentioned first face area.

In another possible implementation, at least one of the first face frames has two face frames with pixel area resolutions greater than the resolution threshold, and let these two face frames be the third face frame and Fourth face frame. Calculate the overlap ratio between the third face frame and the fourth face frame, and in the case that the overlap ratio between the third face frame and the fourth face frame is determined to be less than or equal to the overlap ratio threshold The pixel point area contained in any one of the face frame and the fourth face frame is used as the above-mentioned first face area. For example, there are two face frames whose resolution of the pixel point region included in at least one first face frame is greater than the resolution threshold: face frame b and face frame c. Calculate the overlap rate between face frame b and face frame c. In the case where the overlap ratio between face frame b and face frame c is less than or equal to the overlap ratio threshold, that is, the pixel area contained in face frame b does not overlap with the pixel area contained in face frame c. . Therefore, the pixel area included in the face frame b may be used as the first face area, and the pixel area included in the face frame c may also be used as the first face area. In the case where the overlap ratio between the face frame b and the face frame c is greater than the overlap ratio threshold, that is, the pixel area included in the face frame b and the pixel area included in the face frame c overlap. Therefore, the pixel area included in the face frame b cannot be used as the first face area, and the pixel area included in the face frame c cannot be used as the first face area either.

In another possible implementation, at least one of the first face frames has three face frames with pixel area resolutions greater than the resolution threshold, and the three face frames are set as the third face frame, the third face frame Four face frames and fifth face frames. Calculate the overlap rate between the third face frame and the fourth face frame, the overlap ratio between the third face frame and the fifth face frame, and the overlap between the fourth face frame and the fifth face frame. overlap rate. A possible situation is that the overlap ratio between the third face frame and the fourth face frame is greater than the overlap ratio threshold, and the overlap ratio between the third face frame and the fifth face frame is less than or equal to the overlap ratio threshold, The overlap ratio between the fourth face frame and the fifth face frame is less than or equal to the overlap ratio threshold. That is to say, the pixel area included in the third face frame and the pixel area included in the fourth face frame overlap, and the pixel area included in the fifth face frame and the pixels included in the third face frame overlap. There is no overlapping part of the area, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame cannot be used as the first face area, and the pixel area included in the fourth face frame cannot be used as the first face area, only the pixel area included in the fifth face frame. Can be used as the first face area. In another possible situation, the overlap rate between the third face frame and the fourth face frame is greater than the overlap rate threshold, the overlap rate between the third face frame and the fifth face frame is greater than the overlap rate threshold, and the third face frame is greater than the overlap rate threshold. The overlap ratio between the fourth face frame and the fifth face frame is less than or equal to the overlap ratio threshold. That is to say, the pixel area included in the third face frame and the pixel area included in the fourth face frame overlap, and the pixel area included in the fifth face frame and the pixels included in the third face frame overlap. The area has an overlapping part, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame cannot be used as the first face area, the pixel area included in the fourth face frame cannot be used as the first face area, and the pixel area included in the fifth face frame cannot be used as the first face area. as the first face area. In another possible situation, the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, and the overlap ratio between the third face frame and the fifth face frame is less than or equal to the overlap rate threshold, the overlap rate between the fourth face frame and the fifth face frame is less than or equal to the overlap rate threshold. That is to say, the pixel area contained in the third face frame does not overlap with the pixel area contained in the fourth face frame, and the pixel area contained in the fifth face frame and the pixels contained in the third face frame do not overlap. There is no overlapping part of the area, and the pixel point area contained in the fifth face frame and the pixel point area contained in the fourth face frame have no overlapping part. Therefore, the pixel area included in the third face frame can be used as the first face area, the pixel area included in the fourth face frame can be used as the first face area, and the pixel area included in the fifth face frame can also be used as the first face area. as the first face area.

In a possible implementation method for judging whether the overlap ratio between two face frames is greater than the overlap ratio threshold, it is assumed that the sides of the two face frames are both parallel to the x-axis or the y-axis, that is, to the pixel coordinate system of the above-mentioned image to be processed. The face frame aligned with the horizontal or vertical axis. It is assumed that there are two face frames in the above image to be processed: A face frame and B face frame. When judging that the overlap rate between the face frame A and the face frame B is less than or equal to the overlap rate threshold, it should be understood that the edges of the two face frames cannot overlap. Then, there are four cases where the overlap rate between the B face frame and the A face frame is less than or equal to the overlap ratio threshold: in the first case, the B face frame is above the A face frame; in the second case, The B face frame is below the A face frame; in the third case, the B face frame is to the left of the A face frame; in the fourth case, the B face frame is to the right of the A face frame. Let the coordinates of the upper left corner of the face frame A be p1, the coordinates of the lower right corner be p2, the coordinates of the upper left corner of the face frame B are p3, and the coordinates of the lower right corner are p4. When the B face frame is above the A face frame, the ordinate of p1 is smaller than the ordinate of p4; in the case of the B face frame below the A face frame, the ordinate of p3 is smaller than the ordinate of p2; When the B face frame is to the left of the A face frame, the abscissa of p1 is greater than the abscissa of p4; when the B face frame is to the right of the A face frame, the abscissa of p2 is smaller than that of p3. abscissa. In the above four cases, the overlap rate between the A face frame and the B face frame is less than or equal to the overlap rate threshold. Then, both the pixel point area included in the face frame A and the pixel point area included in the face frame B in the above image to be processed can be used as the first face area.

It should be understood that there are only two relationships between any two different face frames in multiple face frames. One is that the overlap ratio between the two different face frames is greater than the overlap ratio threshold, and the other is that The overlap ratio between two different face frames is less than or equal to the overlap ratio threshold. The embodiments of the present disclosure do not limit the number of multiple face frames.

As an optional implementation manner, when the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, the pixel area included in the third face frame and the fourth face frame The pixel area included in the face frame can be used as the first face area, but it involves the scene of queuing for body temperature detection such as security check. Generally speaking, the closer you are to the temperature measuring device, the larger the corresponding face area will be. . Therefore, the pixel point area contained in the larger face frame in the third face frame and the fourth face frame is taken as the first face area, and the temperature measuring device performs the following steps in the process of executing step 23:

Step 24, taking the pixel area contained in the face frame with the largest size measure in the above-mentioned third face frame and the above-mentioned fourth face frame as the above-mentioned first face area; the size measurement of the above-mentioned third face frame is the above-mentioned No. The maximum value of the side length of the three-person face frame; the size measurement of the above-mentioned fourth face frame is the maximum value of the side length of the above-mentioned fourth face frame;

In some possible implementations, involving queuing for temperature measurement, the front person in the queue shows a larger corresponding face area on the display interface of the temperature measurement device. Therefore, when there are multiple face frames in the image to be processed, the pixel area contained in the largest face frame in the multiple face frames is selected as the first face area. The size of each face frame is measured as the maximum side length of the face frame. Therefore, the maximum value of the side length of each face frame in the multiple face frames in the above image to be processed is obtained first, and then the face frame with the largest size measurement is selected.

In a possible implementation manner, there are two face frames in at least one first face frame, so that the overlap ratio between the two face frames is less than or equal to the overlap ratio threshold, and the two face frames are set as the third person. Face frame and fourth person face frame. When the maximum side length of the third face frame is greater than the maximum side length of the fourth face frame, the pixel area included in the third face frame is used as the first face area. For example, the length of the third face frame is 40 pixels and the width is 45 pixels. Then the size measurement of the third face frame is 45 pixels. The fourth face frame has a length of 36 pixels and a width of 32 pixels. Then the size measurement of the fourth face frame is 36 pixels. Because the size measure of the third face frame is 45 pixels, the size measure of the fourth face frame is 36 pixels. The size measure of the third face frame is greater than the size measure of the fourth face frame, so the pixel area included in the third face frame is used as the first face area.

In another possible implementation manner, the resolution of the pixel area contained in two face frames in at least one first face frame is greater than the resolution threshold, and the overlap ratio between the two face frames is less than or equal to Overlap rate threshold, set the two face frames as the third face frame and the fourth face frame. When the maximum side length of the third face frame is greater than the maximum side length of the fourth face frame, the pixel area included in the third face frame is used as the first face area. For example, the length of the third face frame is 40 pixels and the width is 45 pixels. Then the size measurement of the third face frame is 45 pixels. The fourth face frame has a length of 36 pixels and a width of 32 pixels. Then the size measurement of the fourth face frame is 36 pixels. Because the size measure of the third face frame is 45 pixels, the size measure of the fourth face frame is 36 pixels. The size measure of the third face frame is greater than the size measure of the fourth face frame, so the pixel area included in the third face frame is used as the first face area.

It should be understood that the embodiment of the present disclosure only provides an implementation manner of selecting the largest face frame from two face frames, but this method is also applicable to selecting the largest face frame from the number of face frames greater than 2. The present disclosure The embodiment does not limit the number of face frames. As an optional implementation manner, the temperature measuring device performs the following steps in the process of performing step 103:

Step 31: Based on the above-mentioned homography matrix and the four corner points of the face frame including the above-mentioned first face region, determine four pixel points corresponding to the above-mentioned four corner points from the above-mentioned temperature heat map.

In some possible implementations, the coordinates of the four corners of the face frame of the temperature measurement object in the image to be processed include the coordinates of the first corner, and according to the homography matrix and the coordinates of the four corners, we obtain The coordinates of the corresponding four pixel points in the temperature heat map. Through the above method, the homography matrix obtained by performing square card calibration on each temperature measuring device is a 3x3 matrix. Assume that the coordinates of the four corners of the face frame are the coordinates of the first corner (a1, b1), the coordinates of the second corner (a2, b2), the coordinates of the third corner (a3, b3), and the coordinates of the fourth corner. (a4, b4), through the coordinates of the four corner points and the homography matrix, the coordinates of the four pixel points obtained are the first coordinates (x1, y1), the second coordinates (x2, y2), and the third coordinates. (x3, y3), the fourth coordinate (x4, y4).

It should be understood that the process of obtaining the first coordinates (x1, y1) is similar to the process of obtaining the first coordinates (x1, y1) through the first corner coordinates (a1, b1) and the homography matrix H. Then, through the second corner coordinates (a2, b2) and the homography matrix H, the second coordinates (x2, y2) can be obtained; through the third corner coordinates (a3, b3) and the homography matrix H, we can obtain The third coordinates (x3, y3); the fourth coordinates (x4, y4) can be obtained through the fourth corner coordinates (a4, b4) and the homography matrix H.

Step 32: Determine the second face area based on the quadrilateral area determined by the four pixel points.

In some possible implementations, the above four pixel points are points on the temperature heat map, and a quadrilateral area can be obtained by using these four pixel points as the four vertices of a quadrilateral. Each pixel in the temperature heat map carries the temperature information of the corresponding pixel. Optionally, the temperature heat map is collected by an infrared thermal imaging device on the temperature measuring device. The temperature measuring device obtains a quadrilateral area on the temperature heat map, and uses the quadrilateral area as the second face area, that is, the pixel point area corresponding to the first face area.

As an optional implementation manner, the temperature measuring device performs the following steps in the process of performing step 32:

Obtain the intersection point of the diagonal lines of the quadrilateral area; take the distance between the first point and the intersection point as the first distance; the first point is the pixel point closest to the intersection point among the four pixel points; the intersection point is The center of the circle and the above-mentioned first distance are the radius, and the first area is constructed; the intersection of the above-mentioned first area and the above-mentioned quadrilateral area is determined to obtain the second area;

From the above-mentioned second area, select the area including the above-mentioned intersection as the above-mentioned second face area;

In some possible implementations, the coordinates of the four pixels in the temperature heat map are obtained through the homography matrix and the coordinates of the four corners of the face frame of the RGB image. In the quadrilateral area determined by four pixels in the temperature heat map, if the quadrilateral area is directly used as the second face area, the pixels in the second face area may exist other than those corresponding to the first face area. Pixels corresponding to other areas will cause a decrease in the accuracy of the temperature measurement results of the measurement object. It should be understood that the embodiment of the present disclosure involves a single-person short-range temperature measurement, so the middle part of the quadrilateral area may be a part of the first face area. Therefore, it is necessary to determine an area exactly corresponding to the first face area from the pixel area included in the quadrilateral area, wherein the pixel corresponding to the intersection of the diagonal lines of the quadrilateral area may be a pixel corresponding to the first face area .

Determine the distance between the four pixel points and the intersection point, record the point closest to the intersection point as the first point, and take the distance between the first point and the intersection point as the first distance. The first area is an area with the intersection as the center and the first distance as the radius. However, the pixel area contained in the first area may be larger than the pixel area contained in the quadrilateral. Therefore, a second area is selected from the first area, and the second area is an overlapping area of the pixel point area included in the first area and the quadrilateral area. The second area constructed is the area of the middle part of the quadrangular area.

When the temperature measurement module adopts the temperature of any part of the pixel area corresponding to the first face area in the temperature heat map to obtain the temperature of the temperature measurement object of the above image to be processed, it is necessary to determine this Part of the pixel area has a high probability of being the area corresponding to the first face area. It can be determined that the area in the middle part of the quadrilateral area is the pixel area corresponding to the first face area, that is, the constructed second area is likely to be the pixel area corresponding to the first face area. Selecting an area including the intersection point from the second area as the second face area can improve the accuracy of the temperature measurement result.

In a possible implementation manner, the pixels corresponding to the intersection points are used as the second face region.

In another possible implementation manner, the pixel point area included in the largest inscribed circle of the second area or the pixel point area included in the largest inscribed rectangle is used as the second face area.

In another possible implementation manner, the pixel point area included in the largest inscribed circle with the intersection as the center in the second area is used as the second face area.

As an optional implementation manner, the temperature measuring device performs the following steps in the process of performing step 32:

Obtain the intersection of the diagonal lines of the quadrilateral area; determine the maximum inscribed area of the quadrilateral area; the above-mentioned maximum inscribed area is a rectangular area or a circular area including the above-mentioned intersection point; The above-mentioned largest inscribed area is used as the above-mentioned second face area;

In some possible implementations, the coordinates of the four pixels in the temperature heat map can be obtained through the homography matrix and the coordinates of the four corners of the face frame of the RGB image. The quadrilateral area defined by four pixels in the temperature heatmap. It can be determined that the pixel point corresponding to the intersection of the diagonal lines of the quadrilateral area may be one of the pixel points corresponding to the first face area. Because the embodiment of the present disclosure involves single-person short-range temperature measurement, according to the positional relationship between the RGB imaging device and the infrared imaging device in the temperature measurement device and the mapping relationship of the homography matrix, the largest inscribed rectangle containing the intersection point is intercepted from the quadrilateral area. The area or the largest inscribed circular area can contain the pixel area corresponding to the first face area in the temperature heat map. Through the above homography matrix and the coordinates of the four corner points, the coordinates of the four pixel points are obtained, and then the largest inscribed rectangular area or the largest inscribed circular area is obtained. Then, the largest inscribed rectangular area or the largest inscribed circular area is the second face area in the temperature heat map.

In a possible implementation manner in which the shape of the maximum inscribed area is determined to be a circle, the four pixel points of the quadrilateral area are set as A, B, C, and D, and the intersection point is set as E. Then the four sides of the quadrilateral area are AB, BD, CD, and AC. Calculate the distances from the intersection point E to AB, BD, CD, and AC, respectively. Let the edge closest to the intersection point be AB, and record the distance between AB and the intersection point E as the second distance. Taking the above-mentioned intersection as the center of the circle and the second distance as the radius, the pixel area contained in the circle is the largest inscribed area.

In another possible implementation manner in which the shape of the maximum inscribed area is determined to be a circle, the four pixel points of the quadrilateral area are set as A, B, C, and D, and the intersection point is set as E. Then the four sides of the quadrilateral area are AB, BD, CD, and AC. Find a point A* on AB so that A*E is perpendicular to AB; find a point B* on BD so that B*E is perpendicular to BD; find a point C* on CD so that C*E is perpendicular to CD; find a point on AC D*, such that D*E is perpendicular to AC. The distance between point D* and point B* is taken as the third distance, and the distance between point A* and point C* is taken as the fourth distance. Compare the size of the third distance and the fourth distance, assuming that the third distance is smaller than the fourth distance, take the midpoint of the connection line D*B* corresponding to the third distance as F, and take half of the third distance as the fifth distance. Taking the midpoint F as the center of the circle and the fifth distance as the radius, the pixel area contained in the circle is the largest inscribed area.

In another possible implementation manner in which the shape of the maximum inscribed area is determined to be a circle, the four pixel points of the quadrilateral area are set as A, B, C, and D, and the intersection point is set as E. Then the four sides of the quadrilateral area are AB, BD, CD, and AC. Find a point A* on AB so that A*E is perpendicular to AB; find a point B* on BD so that B*E is perpendicular to BD; find a point C* on CD so that C*E is perpendicular to CD; find a point on AC D*, such that D*E is perpendicular to AC. The distance between point D* and point B* is taken as the third distance, and the distance between point A* and point C* is taken as the fourth distance. Let the intersection of A*C* and B*D* be G, compare the size of the third distance and the fourth distance, assuming that the third distance is less than the fourth distance, take half of the third distance as the fifth distance. Taking the midpoint G as the center of the circle, and the fifth distance as the radius, the pixel area contained in the circle is the largest inscribed area.

In a possible implementation manner in which the shape of the maximum inscribed area is determined to be a rectangle, the four pixels of the quadrilateral area are set as A, B, C, and D. Then the four sides of the quadrilateral area are AB, BD, CD, and AC. Suppose AB is the shortest of the four sides of the quadrilateral area. Then, find a point H on CD so that HA is perpendicular to AB; find a point I on CD so that IB is perpendicular to AB. Then find a little J on BI so that HJ is perpendicular to BI. Because the quadrilateral area enclosed by the four pixels A, B, I, and H has three right angles, the quadrilateral ABIH is a rectangle. The pixel area contained in the rectangle ABIH is the largest inscribed area.

In another possible implementation manner in which the shape of the largest inscribed region is determined to be a rectangle, the four pixel points of the quadrilateral region are set as A, B, C, and D, and the intersection point is set as E. Calculate the lengths of EA, EB, EC and ED, and let the length of EA be the shortest. Find a point K on BE so that the length of EK equals the length of EA; find a point L on DE so that the length of EL equals the length of EA; find a point M on CE so that the length of EM equals the length of EA. That is to say, EK=EA=EL=EM, E is the midpoint of the line segment KM, and point E is also the midpoint of the line segment AL. Because the diagonals of the quadrilateral area enclosed by the four pixels A, K, L, and M are equal and bisected by the point E, the quadrilateral AKLM is a rectangle. The pixel area contained in the rectangle AKLM is the largest inscribed area.

In another possible implementation manner in which the shape of the maximum inscribed area is determined to be a rectangle, the four pixels of the quadrilateral area are set as A, B, C, and D. Then the four sides of the quadrilateral area are AB, BD, CD, and AC. Take the midpoint of AB as N; take the midpoint of BD as O; take the midpoint of CD as P; take the midpoint of AC as R. Because the quadrilateral enclosed by the midpoints of any quadrilateral is a parallelogram. That is, the quadrilateral NOPR is a parallelogram. Among them, PN and OR are the two diagonals corresponding to the parallelogram NOPR, and the intersection of the two diagonals is S. Take the shorter diagonal of the PN line segment and the OR line segment. For example, when the length of PN is less than the length of OR, PN is taken as the diagonal of the rectangular area, where the length of NS is equal to the length of SP. Find a point T on SO, so that the length of TS is equal to the length of NS; find a point U on SR, so that the length of SU is equal to the length of NS. That is to say, NS=SP=TS=SU, S is the midpoint of the line segment PN, and point S is also the midpoint of the line segment UT. Because the diagonals of the quadrilateral area enclosed by the four pixels N, P, T, and U are equal and bisected by the point S, the quadrilateral NPTU is a rectangle. The pixel area contained in the rectangular NPTU is the largest inscribed area.

As an optional implementation manner, in order to obtain an accurate temperature of the temperature measurement object, it is necessary to accurately find the second face region in the temperature heat map. Therefore, the temperature measuring device performs the following steps in the process of determining the maximum inscribed area of the above quadrilateral area:

Select the second largest abscissa from the abscissas of the above four pixels to obtain the first abscissa; select the third largest abscissa from the abscissas of the above four pixels to obtain the second abscissa; from the above Select the second largest ordinate from the ordinates of the four pixel points to obtain the first ordinate; select the third largest ordinate from the ordinates of the above four pixels to obtain the second ordinate; The abscissa and the first ordinate determine the second point; the third point is determined based on the first abscissa and the second ordinate; the fourth point is determined based on the second abscissa and the first ordinate; The second abscissa and the second ordinate determine the fifth point; the area determined by the second point, the third point, the fourth point and the fifth point is used as the maximum inscribed area;

In some possible implementations, the four pixel points are set as A, B, C, and D, respectively. Among them, the coordinates of A are (X1, Y1); the coordinates of B are (X2, Y2); the coordinates of C are (X3, Y3); the coordinates of D are (X4, Y4). Sort X1, X2, X3, X4, sort Y1, Y2, Y3, Y4. Assuming X1>X2>X3>X4, then the first abscissa is X2, and the second abscissa is X3. Assuming that Y1>Y2>Y3>Y4, then the first ordinate is Y2, and the second ordinate is Y3. The coordinates of the second point are (X2, Y2), the coordinates of the third point are (X2, Y3), the coordinates of the fourth point are (X3, Y2), and the coordinates of the fifth point are (X3, Y3). According to the coordinate relationship of the second point, the third point, the fourth point and the fifth point, it can be determined that the quadrilateral area obtained by sequentially connecting the second point, the third point, the fourth point and the fifth point is a rectangular area. Let the rectangular area obtained above be the above-mentioned maximum inscribed area. As an optional implementation manner, in order to further improve the accuracy of temperature measurement, the temperature measurement device uses the temperature of the forehead region of the first face region as the temperature of the temperature measurement object, and the temperature measurement device executes in the process of performing step 104 The following steps:

Step 41: Perform forehead detection on the above-mentioned first face area, and obtain the forehead detection result of the above-mentioned first face area;

In some possible implementation manners, performing forehead detection on the first face area, and obtaining a detection result includes: the forehead of the person in the first face area is in a blocked state or the forehead of the person in the first face area is in the unblocked state. occlusion state.

In some possible implementations, the temperature measurement device performs a first feature extraction process on the first face region to obtain first feature data, wherein the first feature data carries whether the forehead of the person in the first face region is blocked status information. The temperature measuring device obtains the detection result based on the first characteristic data obtained by the forehead detection.

Optionally, the first feature extraction process may be implemented by a forehead detection network. A forehead detection network can be obtained by training a deep convolutional neural network by using at least one first training image with label information as training data, wherein the label information includes whether the forehead of the person in the first training image is occluded. state.

The embodiments of the present disclosure do not limit the manner used for forehead detection.

Step 42: In the case where the forehead detection result of the first face region is that the forehead region in the first face region is not blocked, obtain the temperature measurement based on the temperature of the pixels included in the third face region. The temperature of the object; the above-mentioned third face area is the area corresponding to the above-mentioned forehead area in the above-mentioned second face area;

In some possible implementations, when the forehead detection result of the first face region is that the forehead region in the first face region is not blocked, first find the third face region from the temperature heat map . The third face area is a pixel area in the second face area that corresponds to the forehead area in the first face area. Generally speaking, the forehead area is located in the upper 30% to 40% of the entire face area. That is to say, the third face area is the upper 30% to 40% of the second face area. Based on the temperature of the pixels included in the third face region in the temperature heat map, the temperature of the temperature measurement object corresponding to the first face region is obtained. In the case that the forehead detection result of the first face region is that the forehead region in the first face region is blocked, prompt information that the forehead needs to be exposed is output.

In a possible implementation manner, the temperature measurement device is placed indoors, and the average temperature of the pixel area of the third face area is used as the temperature of the temperature measurement object. For example, the pixel points of the third face region include: pixel point a and pixel point b. Among them, the temperature corresponding to the pixel point a is 36.9 degrees, and the temperature corresponding to the pixel point b is 36.3 degrees. The temperature measurement device can take the average value (36.6 degrees) of the temperature corresponding to the pixel point a and the temperature corresponding to the pixel point b as the temperature of the temperature measurement object.

In another possible implementation manner, the temperature measuring device uses the highest temperature value corresponding to the pixel point in the third face region as the temperature of the temperature measuring object. For example, the pixels in the third face area include: pixel a, pixel b, and pixel c, where the temperature corresponding to pixel a is 36.9 degrees, the temperature corresponding to pixel b is 36.3 degrees, and the temperature corresponding to pixel a is 36.3 degrees. c corresponds to a temperature of 37 degrees. The temperature measurement device can take the temperature corresponding to the pixel point a, the temperature corresponding to the pixel point b, and the maximum value (37 degrees) of the temperature corresponding to the pixel point c as the temperature of the temperature measurement object.

In another possible implementation manner, the temperature measuring device is placed outdoors and may be exposed to direct sunlight. And because the third face area may contain pixel area corresponding to the direct sun area. If the temperature of the pixel point corresponding to the direct sunlight area and the temperature of the actual face area are averaged as the temperature of the temperature measurement object in the above image to be processed, there will be a relatively large error. The temperature of direct sunlight is about 50 degrees, and the temperature that the human body can tolerate is generally below 45 degrees. Therefore, read the temperature of each pixel in the pixel area of the third face area, exclude the pixels whose temperature is higher than 45 degrees, and average the temperature of the pixels whose temperature is lower than 45 degrees as the above-mentioned pending processing The temperature of the thermometric object in the image.

It should be understood that, in the embodiment of the present disclosure, the average value of the temperature of a part of the pixel points included in the third face area may be used as the temperature of the temperature measurement object, or the temperature of a part of the pixel points included in the third face area may be used. The highest value of the temperature is used as the temperature of the temperature measurement object, which is not limited in the embodiment of the present disclosure.

Those skilled in the art can understand that, in the methods provided by the above embodiments, the writing order of each step does not mean a strict execution order but constitutes any limitation on the implementation process, and the execution order of each step should be based on its functions and possible Internal logic is determined.

The methods of the embodiments of the present disclosure are described in detail above, and the apparatuses of the embodiments of the present disclosure are provided below.

Please refer to FIG. 2. FIG. 2 is a schematic structural diagram of a temperature measurement device according to an embodiment of the present disclosure. The temperature measurement device 1 includes: an acquisition part 11, a detection part 12, a first processing part 13, and a second processing part 14, in:

an acquisition part 11, configured to acquire an image to be processed, a temperature heat map, and a homography matrix between the temperature heat map and the to-be-processed image;

The detection part 12 is configured to perform face detection on the to-be-processed image to obtain a first face area;

The first processing part 13 is configured to determine the pixel area corresponding to the first face area from the temperature heat map based on the homography matrix to obtain a second face area;

The second processing part 14 is configured to determine the temperature of the temperature measurement object corresponding to the first face area based on the temperature of the pixels included in the second face frame.

In some possible implementation manners, the first processing part 13 is further configured to:

Based on the homography matrix and the four corner points of the face frame including the first face region, determine four pixel points corresponding to the four corner points from the temperature heat map;

The second face area is determined based on the quadrilateral area determined by the four pixel points.

In some possible implementations, the detection part 12 is further configured to:

performing face detection on the to-be-processed image to obtain at least one first face frame;

The pixel area included in the second face frame in the at least one first face frame is used as the first face area; the resolution of the pixel area included in the second face frame is greater than the resolution threshold face frame.

In some possible implementations, the number of the first face frames is greater than 1, and the at least one first face frame includes a third face frame and a fourth face frame;

The detection part 12 is also configured to:

In the case that the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, any one of the third face frame and the fourth face frame is The pixel area included in the face frame is used as the first face area.

In some possible implementations, the detection part 12 is further configured to:

The pixel area contained in the face frame with the largest size measure in the third face frame and the fourth face frame is taken as the first face area; the size measure of the third face frame is The maximum value of the side length of the third face frame; the size measurement of the fourth face frame is the maximum value of the side length of the fourth face frame.

In some possible implementation manners, the second processing part 14 is further configured to:

performing forehead detection on the first face region to obtain a forehead detection result of the first face region;

When the forehead detection result of the first face area is that the forehead area in the first face area is not blocked, the temperature measurement is obtained based on the temperature of the pixels included in the third face area The temperature of the object; the third face region is the region corresponding to the forehead region in the second face region.

In some possible implementation manners, the first processing part 13 is further configured to: obtain the intersection of the diagonal lines of the quadrilateral area; take the distance between the first point and the intersection as the first distance; The first point is the pixel point closest to the intersection point among the four pixel points; taking the intersection point as the center of the circle and the first distance as the radius, a first area is constructed; The intersection of the quadrilateral areas is obtained to obtain a second area; the area including the intersection is selected from the second area as the second face area.

In some possible implementation manners, the first processing part 13 is further configured to: obtain the intersection of the diagonal lines of the quadrilateral area; determine the maximum inscribed area of the quadrilateral area; the maximum inscribed area is a rectangular area or a circular area including the intersection point; the largest inscribed area is taken as the second face area.

In some possible implementation manners, the first processing part 13 is further configured to: select the second largest abscissa from the abscissas of the four pixel points to obtain the first abscissa, and obtain the first abscissa from the abscissas of the four pixels. From the abscissa of the pixel point, select the third largest abscissa to obtain the second abscissa, select the second largest ordinate from the ordinate of the four pixel points to obtain the first ordinate, and obtain the first ordinate from the four pixel points. Select the third largest ordinate to obtain the second ordinate; determine the second point based on the first abscissa and the first ordinate, and determine the second point based on the first abscissa and the second ordinate A third point is determined, a fourth point is determined based on the second abscissa and the first ordinate, and a fifth point is determined based on the second abscissa and the second ordinate; the second point, The area determined by the third point, the fourth point and the fifth point is used as the maximum inscribed area.

In some possible implementations, the temperature measuring device can accurately determine the face region of the temperature measurement object from the to-be-processed image by performing face detection on the to-be-processed image to obtain the first face region. Based on the homography matrix between the image to be processed and the temperature heat map and the position of the first face region in the image to be processed, determining the face region of the temperature measurement object from the temperature heat map can improve the performance from the temperature heat map. Determine the accuracy of the face area of the temperature measurement object to obtain a second face area. In this way, the temperature measurement device obtains the temperature of the temperature measurement object based on the temperature of the second face region, which can improve the accuracy of the temperature measurement object.

In some possible implementation manners, the functions or included parts of the apparatus provided in the embodiments of the present disclosure may be configured to execute the methods described in the above method embodiments, and for implementation, reference may be made to the descriptions in the above method embodiments, for the sake of brevity , which will not be repeated here.

FIG. 3 is a schematic diagram of a hardware structure of a temperature measuring device according to an embodiment of the present disclosure. The temperature measuring device 2 includes a processor 21 , a memory 22 , an input device 23 , and an output device 24 . The processor 21, the memory 22, the input device 23, and the output device 24 are coupled through a connector, and the connector includes various types of interfaces, transmission lines, or buses, etc., which are not limited in this embodiment of the present disclosure. It should be understood that, in various embodiments of the present disclosure, coupling refers to mutual connection in a specific manner, including direct connection or indirect connection through other devices, such as various interfaces, transmission lines, and buses.

The processor 21 may be one or more graphics processing units (graphics processing units, GPUs). In the case where the processor 21 is a GPU, the GPU may be a single-core GPU or a multi-core GPU. Optionally, the processor 21 may be a processor group composed of multiple GPUs, and the multiple processors are coupled to each other through one or more buses. Optionally, the processor may also be another type of processor, etc., which is not limited in this embodiment of the present disclosure.

The memory 22 may be configured to store computer program instructions, as well as various types of computer program code, including program code configured to execute aspects of the embodiments of the present disclosure. Optionally, the memory includes, but is not limited to, random access memory (RAM), read-only memory (read-only memory, ROM), erasable programmable read-only memory (erasable programmable read only memory, EPROM) ), or a portable disc read-only memory (CD-ROM), which is configured for associated instructions and data.

The input device 23 is configured to input data and/or signals, and the output device 24 is configured to output data and/or signals. The input device 23 and the output device 24 may be independent devices or may be an integral device.

It can be understood that, in some possible implementation manners, the memory 22 can not only store relevant instructions, but also store relevant data. For example, the memory 22 can store the image to be processed and the temperature heat map obtained through the input device 23 , or the memory 22 The temperature of the temperature measurement object obtained through the processor 21 and the like may also be stored, and the data stored in the memory is not limited in this embodiment of the present disclosure.

It can be understood that FIG. 3 only shows a simplified design of the temperature measuring device. In practical applications, the temperature measurement device may also include other necessary components, including but not limited to any number of input/output devices, processors, memories, etc., and all temperature measurement devices that can implement the embodiments of the present disclosure are included in this disclosure. within the scope of the disclosed embodiments.

Those of ordinary skill in the art can realize that the parts and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the disclosed embodiments.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the working process of the above-described system, device and part, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here. Those skilled in the art can also clearly understand that the description of each embodiment of the present disclosure has its own emphasis. For the convenience and brevity of the description, the same or similar parts may not be repeated in different embodiments. Therefore, in a certain embodiment For the parts that are not described or not described in detail, reference may be made to the descriptions of other embodiments.

In the several embodiments provided by the present disclosure, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the parts is only a logical function division, and there may be other division methods in the actual implementation process, for example, multiple parts or components may be divided into Incorporation may either be integrated into another system, or some features may be omitted, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or parts, and may be in electrical, mechanical or other forms.

The parts described as separate parts may or may not be physically separated, and the parts shown as parts may or may not be physical parts, that is, they may be located in one place, or may be distributed over multiple network parts. Some or all of them may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional part in each embodiment of the present disclosure may be integrated into one processing part, or each part may exist physically alone, or two or more parts may be integrated into one part.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. In the case of implementation in software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the procedures or functions described in accordance with the embodiments of the present disclosure are produced in whole or in part. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted over a computer-readable storage medium. The computer instructions can be sent from a website site, computer, server, or data center via wired (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.) another website site, computer, server or data center for transmission. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media. The available media may be magnetic media (eg, floppy disk, hard disk, magnetic tape), optical media (eg, digital versatile disc (DVD)), or semiconductor media (eg, solid state disk (SSD) ))Wait.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented, and the process can be completed by instructing the relevant hardware by a computer program, and the program can be stored in a computer-readable storage medium. The process may include the flow of each method embodiment described above. The aforementioned storage medium includes: read-only memory (read-only memory, ROM) or random access memory (random access memory, RAM), magnetic disk or optical disk and other media that can store program codes.

Industrial Applicability

Embodiments of the present disclosure provide a temperature measurement method and device, an electronic device, and a storage medium. The method includes: acquiring an image to be processed, a temperature heat map, and a homography matrix between the temperature heat map and the to-be-processed image; performing face detection on the to-be-processed image to obtain a first face region; The pixel area corresponding to the first face area is determined from the temperature heat map based on the homography matrix to obtain a second face area; based on the temperature of the pixels included in the second face area , and determine the temperature of the temperature measurement object corresponding to the first face region. It can accurately determine the face area of the temperature measurement object from the image to be processed to obtain the first face area, improve the accuracy of the temperature measurement object's face area determined from the temperature thermogram, and then improve the temperature of the temperature measurement object. accuracy.

Claims (14)

1. A temperature measurement method, the method comprising:

   acquiring an image to be processed, a temperature heat map, and a homography matrix between the temperature heat map and the to-be-processed image;

   performing face detection on the to-be-processed image to obtain a first face region;

   Determine the pixel area corresponding to the first face area from the temperature heat map based on the homography matrix to obtain a second face area;

   Based on the temperature of the pixel points included in the second face area, the temperature of the temperature measurement object corresponding to the first face area is determined.
2. The method according to claim 1, wherein the determining a pixel region corresponding to the first face region from the temperature heat map based on the homography matrix to obtain a second face region, comprising:

   Based on the homography matrix and the four corner points of the face frame including the first face region, determine four pixel points corresponding to the four corner points from the temperature heat map;

   The second face area is determined based on the quadrilateral area determined by the four pixel points.
3. The method according to claim 1 or 2, wherein the performing face detection on the to-be-processed image to obtain the first face region comprises:

   performing face detection on the to-be-processed image to obtain at least one first face frame;

   The pixel point area included in the second face frame in the at least one first face frame is used as the first face area; the second face frame is the at least one first face frame contained in the first face frame The resolution of the pixel area is greater than the resolution threshold of the face frame.
4. The method of claim 3, wherein the number of the first face frames is greater than 1, and the at least one first face frame includes a third face frame and a fourth face frame;

   The pixel point area included in the second face frame in the at least one first face frame is used as the first face area, including:

   In the case that the overlap ratio between the third face frame and the fourth face frame is less than or equal to the overlap ratio threshold, any one of the third face frame and the fourth face frame is The pixel area included in the face frame is used as the first face area.
5. The method according to claim 4, wherein the pixel point area included in any one of the third face frame and the fourth face frame as the first face area, comprising: :

   The pixel area contained in the face frame with the largest size measure in the third face frame and the fourth face frame is taken as the first face area; the size measure of the third face frame is The maximum value of the side length of the third face frame; the size measurement of the fourth face frame is the maximum value of the side length of the fourth face frame.
6. The method according to any one of claims 1 to 5, wherein the determining the temperature of the temperature measurement object corresponding to the first face region based on the temperature of the pixel points included in the second face region comprises: :

   performing forehead detection on the first face region to obtain a forehead detection result of the first face region;

   When the forehead detection result of the first face area is that the forehead area in the first face area is not blocked, the temperature measurement is obtained based on the temperature of the pixels included in the third face area The temperature of the object; the third face region is the region corresponding to the forehead region in the second face region.
7. The method according to any one of claims 2 to 6, wherein the determining the second face region based on the quadrilateral region determined by the four pixel points comprises:

   obtain the intersection of the diagonals of the quadrilateral area;

   Taking the distance between the first point and the intersection point as the first distance; the first point is the pixel point closest to the intersection point among the four pixel points;

   Constructing the first region with the intersection as the center and the first distance as the radius;

   determining the intersection of the first area and the quadrilateral area to obtain a second area;

   A region including the intersection point is selected from the second region as the second face region.
8. The method according to any one of claims 2 to 6, wherein the determining the second face region based on the quadrilateral region determined by the four pixel points comprises:

   obtain the intersection of the diagonals of the quadrilateral area;

   determining the largest inscribed area of the quadrilateral area; the largest inscribed area is a rectangular area or a circular area including the intersection;

   The largest inscribed area is used as the second face area.
9. 9. The method of claim 8, wherein the determining a maximum inscribed area of the quadrilateral area comprises:

   Select the second largest abscissa from the abscissas of the four pixels to obtain the first abscissa, select the third largest abscissa from the abscissas of the four pixels to obtain the second abscissa, and obtain the second abscissa from all the abscissas. From the vertical coordinates of the four pixel points, the second largest vertical coordinate is selected to obtain the first vertical coordinate, and the third largest vertical coordinate is selected from the vertical coordinates of the four pixel points to obtain the second vertical coordinate;

   A second point is determined based on the first abscissa and the first ordinate, a third point is determined based on the first abscissa and the second ordinate, and a third point is determined based on the second abscissa and the first The ordinate determines the fourth point, and the fifth point is determined based on the second abscissa and the second ordinate;

   The area determined by the second point, the third point, the fourth point and the fifth point is taken as the maximum inscribed area.
10. A temperature measuring device comprising:

    an acquisition part, configured to acquire an image to be processed, a temperature heat map, and a homography matrix between the temperature heat map and the to-be-processed image;

    a detection part, configured to perform face detection on the to-be-processed image to obtain a first face region;

    a first processing part, configured to determine a pixel area corresponding to the first face area from the temperature heat map based on the homography matrix to obtain a second face area;

    The second processing part is configured to determine the temperature of the temperature measurement object corresponding to the first face area based on the temperature of the pixels included in the second face frame.
11. A processor configured to perform the method of any of claims 1 to 9.
12. An electronic device comprising: a processor and a memory, the memory configured to store computer program code, the computer program code comprising computer instructions, the electronic device when the processor executes the computer instructions A method as claimed in any one of claims 1 to 9 is performed.
13. A computer-readable storage medium storing a computer program in the computer-readable storage medium, the computer program comprising program instructions, which, when the program instructions are executed by a processor, cause the processor to execute the claims The method of any one of 1 to 9.
14. A computer program, comprising computer-readable codes, when the computer-readable codes are run in an electronic device, a processor in the electronic device executes the code for realizing any one of claims 1 to 9 Methods.

Images