WO2022121189A1

WO2022121189A1 - Method and apparatus for measuring temperature, and computer device

Info

Publication number: WO2022121189A1
Application number: PCT/CN2021/084571
Authority: WO
Inventors: 孙奥兰; 王健宗; 程宁
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-12-11
Filing date: 2021-03-31
Publication date: 2022-06-16
Also published as: CN112539843A; CN112539843B

Abstract

The present application relates to the field of computer vision analysis of artificial intelligence. Disclosed is a method for measuring a temperature, the method comprising: acquiring a pseudo-color infrared image obtained during an infrared detection process; intercepting a temperature data area in the pseudo-color infrared image; segmenting the temperature data area according to a preset segmentation manner to obtain sub-pictures respectively corresponding to a highest-temperature area and a lowest-temperature area; comparing the sub-pictures with templates in a preset template library to respectively determine a specified template with the highest similarity to each sub-picture; and obtaining, according to a digit of the specified template, a temperature range corresponding to the temperature data area. On the basis that the position of a temperature strip in the pseudo-color infrared image is relatively fixed, and that identification fonts of the highest-temperature area and the lowest-temperature area are also consistent, the sub-pictures of the highest-temperature area and the lowest-temperature area are intercepted from the temperature data area of the temperature strip, and a simple and efficient digital template matching method is used, such that a temperature range in the temperature strip in the pseudo-color infrared image can be obtained; the workload is small; and the recognition efficiency is high.

Description

Method, device and computer equipment for detecting temperature

This application claims the priority of the Chinese patent application with the application number 202011461798.1 and the invention titled "Method, Apparatus and Computer Equipment for Detecting Temperature" filed with the China Patent Office on December 11, 2020, the entire contents of which are incorporated herein by reference Applying.

technical field

This application relates to the field of computer vision analysis of artificial intelligence.

Background technique

In the original infrared image obtained by the infrared thermal imager, the temperature data area displayed is used to record the relevant data of the infrared radiation temperature of the background and the target, not the direct temperature value, and the storage format of the original infrared image is quite special. Data analysis can only be performed after it is processed into a bitmap that can be processed by a machine through professional software. The image obtained after being processed by professional software is a pseudo-color infrared image, which can only characterize the temperature difference of the picture by grayscale and heat map, and cannot obtain specific temperature data information.

The inventor realized that the existing temperature data obtained by recognizing the temperature bar on the pseudo-color infrared image are mostly recognized by the infrared image temperature width recognition method based on deep learning. Classification and recognition, but the workload of preparing training data sets in the early stage is large and the efficiency is low.

technical problem

The existing technical problems of identifying temperature data on a pseudo-color infrared image are heavy workload and low efficiency.

technical solutions

A first aspect of the present application proposes a method for detecting temperature, comprising:

Obtain the pseudo-color infrared image obtained during the infrared detection process;

intercepting the temperature data area in the pseudo-color infrared image;

The sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

Comparing the sub-pictures with the templates in the preset template library, respectively determining the specified template with the highest similarity with each of the sub-pictures;

According to the number of the specified template, the temperature range corresponding to the temperature data area is obtained.

A second aspect of the present application provides a device for detecting temperature, comprising:

a first acquisition module, used for acquiring a pseudo-color infrared image obtained during the infrared detection process;

an interception module, used for intercepting the temperature data area in the pseudo-color infrared image;

a segmentation module, configured to segment the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

a comparison module, configured to compare the sub-pictures with the templates in the preset template library, and respectively determine the specified template with the highest similarity with each of the sub-pictures;

The obtaining module is used to obtain the temperature range corresponding to the temperature data area according to the number of the specified template.

A third aspect of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above method for detecting temperature when the computer program is executed, including: acquiring infrared detection The pseudo-color infrared image obtained in the process; intercepting the temperature data area in the pseudo-color infrared image; dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein, Only one number is included in the sub-picture; the sub-picture is compared with the templates in the preset template library, and the specified template with the highest similarity with each of the sub-pictures is respectively determined; according to the number of the specified template, Obtain the temperature range corresponding to the temperature data area. A fourth aspect of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned method for detecting temperature includes: acquiring a false temperature obtained during infrared detection. color infrared image; intercepting the temperature data area in the pseudo-color infrared image; dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein, in the sub-pictures Only one number is included; the sub-picture is compared with the templates in the preset template library, and the specified template with the highest similarity with each of the sub-pictures is respectively determined; according to the number of the specified template, the temperature data is obtained The temperature range corresponding to the region.

beneficial effect

This application is based on the fact that the position of the temperature bar in the pseudo-color infrared image is relatively fixed, and the identification fonts of the highest temperature area and the lowest temperature area above and below the temperature bar are also consistent. By cutting out the highest temperature from the temperature data area of the temperature bar The sub-picture of the region and the lowest temperature region, and using a simple and efficient digital template matching method, the temperature range in the temperature bar in the pseudo-color infrared image can be obtained, with a small workload and high recognition efficiency.

Description of drawings

1 is a schematic flowchart of a method for detecting temperature according to an embodiment of the present application;

2 is a schematic diagram of a device for detecting temperature according to an embodiment of the present application;

FIG. 3 is a schematic diagram of an internal structure of a computer device according to an embodiment of the present application.

Embodiments of the present invention

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

1 , a method for detecting temperature according to an embodiment of the present application includes:

S1: Acquire the pseudo-color infrared image obtained in the infrared detection process;

S2: Intercept the temperature data area in the pseudo-color infrared image;

S3: Divide the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein the sub-picture includes only one number;

S4: Comparing the sub-picture with the templates in the preset template library, and respectively determining the specified template with the highest similarity with each of the sub-pictures;

S5: Obtain the temperature range corresponding to the temperature data area according to the number of the specified template.

In the embodiment of the present application, only the numerical labels corresponding to the highest temperature and the lowest temperature are displayed in the temperature bar, and the other areas of the temperature bar are indicated by scales. By identifying the digital identification fonts of the highest temperature area and the lowest temperature area, the temperature data area is determined and cut out by vertical projection. The above-mentioned preset segmentation method includes a segmentation method in which the boundary is determined by the gray scale change rule. For example, the background grayscale in the above pseudo-color infrared image is different from the object imaging grayscale, and the pseudo-color infrared image detected by infrared detection follows the difference in temperature, and the grayscale is also different. Area grayscale is large. The temperature range is obtained by intercepting a sub-picture containing only one digital image, then determining the number of the sub-picture according to the template, and obtaining the highest temperature and the lowest temperature according to the positional relationship of the sub-picture.

The above-mentioned preset template library includes a binarized image template of ten numbers "0-9", and a binary image template of "minus sign". Comparing the captured sub-picture containing only one number with all templates in the preset template library one by one, find the template that is most similar to the sub-picture in the preset template library, and use it as the identification result of the sub-picture. For example, it is recognized that there are two sub-pictures from left to right in the highest temperature area. By comparing the digital content of the left sub-picture is 8 and the right digital content is 5, the highest temperature is 85 degrees. When comparing the sub-picture and the template in the embodiment of the present application, the comparison is performed by calculating the similarity between the two.

The examples of this application use SSIM (Structural SIMilarity, structural similarity), and the range of the comparison result is [-1, 1]. The closer the comparison result is to -1, the greater the difference between the two, and the closer the comparison result is to 1. The more similar the two are. In addition, SSIM(x,y)=[l(x,y)] ^α [c(x,y)] ^β [s(x,y)] ^γ , where x and y are the input two images to be compared , α>0, β>0, γ>0, through

Make brightness comparisons; pass

Make a contrast comparison; pass

For structural comparison, μ _x and μ _y represent the grayscale mean of _x and _y , respectively, σx and σy represent the grayscale standard deviation of x and y, respectively, _σxy represent the grayscale covariance of x and y, and c ₁ , c ₂ , and c ₃ are constants, respectively, to avoid system errors when the denominator is 0. In the embodiments of the present application, according to the comparative experimental data of the pseudo-color infrared images, α=β=γ=1, and c ₃ =c ₂ /2 are set to simplify the calculation. The embodiment of the present application simplifies the SSIM as follows:

The larger the SSIM value, the smaller the difference between the two, that is, when the two contrast images are exactly the same, SSIM=1.

Further, the temperature data area also includes a grayscale range, and the step S3 of dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area from the temperature data area according to a preset segmentation method includes:

S31: Determine the current grayscale threshold within the grayscale range corresponding to the temperature data area;

S32: Divide the grayscale range of the temperature data region into two regions by the current grayscale threshold, including a first region smaller than the current grayscale threshold and a second region greater than or equal to the current grayscale threshold;

S33: Calculate the grayscale variance of the first area and the second area under the current grayscale threshold;

S34: Dynamically adjust the grayscale threshold within the grayscale range corresponding to the temperature data region, and determine the maximum grayscale variance corresponding to the first region and the second region;

S35: Use the grayscale threshold corresponding to the maximum grayscale variance as the segmentation threshold;

S36: Segment the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively according to the segmentation threshold.

In the embodiment of the present application, the gray threshold corresponding to the maximum gray variance is used as the segmentation threshold of the boundary, so that the selected segmentation threshold can distinguish the average gray level of the scene area, the average gray level of the background area, and the entire pseudo-color infrared image. The difference between the average gray levels within is the largest. The embodiment of the present application determines the segmentation threshold by traversing the grayscale values of the entire grayscale range of the pseudo-color infrared image, and calculating the maximum grayscale variance of the two pixel regions of the temperature data region divided by the currently selected grayscale value.

Further, the step S33 of calculating the grayscale variance of the first area and the second area under the current grayscale threshold includes:

S331: Calculate the first probability of each gray value distribution in the first area, and calculate the second probability of each gray value distribution in the second area;

S332: Calculate a first average grayscale value corresponding to the first region according to each of the first probabilities, and calculate a second average grayscale value corresponding to the second region according to each of the second probabilities;

S333: Calculate the total average grayscale of the temperature data region according to the first average grayscale value and the second average grayscale value;

S334: Calculate the first region and the The grayscale variance of the two regions under the current grayscale threshold.

For example, the number of pixels with grayscale i in the temperature data area is n _i , the grayscale range is [0,L-1], and the total number of pixels is

The probability of occurrence of each gray value is

For _Pi ,

The pixels in the image are divided into two categories or two regions by the grayscale threshold T, such as the first region C ₀ and the second region C ₁ . C ₀ is composed of pixels with gray value [0, T-1], and C ₁ is composed of pixels with gray value in [T, L-1], then the first probability corresponding to C ₀ is

The second probability corresponding to _C1 is

The first average gray value corresponding to C ₀ is

The second average gray value corresponding to C ₁ is

Overall average grayscale of the temperature data area

The grayscale variance of the two regions is:

By letting T take values in order within [0, L-1], the T value with the largest grayscale variance is the best segmentation threshold, that is, the sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented.

Further, the step S36 of dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively by the segmentation threshold includes:

S361: Determine the first boundary between the temperature data area and the background area according to the segmentation threshold, as well as the second boundary where the highest temperature digit in the temperature data area is located, and the lowest temperature digit in the temperature data area. third boundary;

S362: Determine the boundary of the sub-picture corresponding to the highest temperature region according to the first boundary and the second boundary, and determine the sub-picture corresponding to the lowest temperature region according to the first boundary and the third boundary boundary;

S363: Intercept the sub-picture corresponding to the highest temperature area according to the boundary of the sub-picture corresponding to the highest temperature area, and intercept the sub-picture corresponding to the lowest temperature area according to the boundary of the sub-picture corresponding to the lowest temperature area.

In the embodiment of the present application, when the temperature data area is taken as the research object, the pixels outside the temperature data area in the pseudo-color infrared image are background areas, and the above-mentioned first boundary is the boundary of the temperature data area, such as a rectangular frame. The second boundary and the third boundary are distinguishing boundaries within the temperature data area, and are used to distinguish areas where numbers exist within the temperature data area. For example, the temperature data area is a vertical rectangular frame in a pseudo-color infrared image, and the second boundary and the third boundary are horizontal boundary lines inside the temperature data area. Then, a sub-picture corresponding to the highest temperature area is enclosed by a part of the vertical rectangular frame and the second boundary, and the sub-picture corresponding to the highest temperature area includes one or more sub-pictures arranged in parallel. A sub-picture corresponding to the lowest temperature region is enclosed by a part of the vertical rectangular frame and the third boundary, and the sub-picture corresponding to the lowest temperature region includes one or more sub-pictures arranged in parallel. Only one number is included in each word picture above.

Further, the second boundary and the third boundary are in the lateral direction of the temperature data area, and the first boundary between the temperature data area and the background area is determined according to the segmentation threshold, and the temperature After the step S361 of the second boundary where the highest temperature number is located in the data area and the third boundary where the lowest temperature number in the temperature data area is located, it includes:

S3611: Determine whether the specified pixel point on the second boundary is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary;

S3612: If it is not within the boundary threshold range of the grayscale change, search for the grayscale fusion point in the longitudinal direction passing through the designated pixel point and parallel to the temperature data region;

S3613: Replace the gray-scale fusion point with the specified pixel point on the second boundary;

S3614: Correct all the pixels on the second boundary according to the correction method of the specified pixel point, and correct the third boundary according to the correction method of the second boundary.

In the embodiment of the present application, since the boundary of the grayscale change in the temperature data area is weak, in order to ensure more accurate segmentation, the horizontal second boundary and the third boundary are corrected by means of watershed judgment to avoid noise in the image. , Over-segmentation caused by subtle grayscale changes on the object surface to ensure the integrity of the digital contours within the sub-picture. The above-mentioned gray-scale fusion point refers to a cliff point where the gray-scale value suddenly increases or decreases suddenly. For example, obtain the grayscale values of several adjacent pixel points in the longitudinal direction parallel to the temperature data area passing through the specified pixel point, calculate the grayscale difference between two adjacent points in turn, and the point where the grayscale difference jumps It is the gray-scale fusion point.

Further, after the step S5 of obtaining the temperature range corresponding to the temperature data area according to the number of the specified template, the step includes:

S6: Obtain the minimum pixel value within the highest temperature region and the maximum pixel value within the lowest temperature region;

S7: Calculate the linear correlation coefficient between the pixel value and the temperature value according to the corresponding relationship between the maximum temperature and the minimum pixel value, and the corresponding relationship between the minimum temperature and the maximum pixel value;

S8: Estimate the temperature values corresponding to all the pixel points in the pseudo-color infrared image according to the linear correlation coefficient.

In the embodiment of the present application, the temperature values corresponding to the pixel values in the temperature bar are linearly distributed, and the corresponding relationship between the maximum temperature and the minimum pixel value and the corresponding relationship between the minimum temperature and the maximum pixel value are obtained. Correlation coefficients, so that the temperature values corresponding to different pixel values in the entire temperature bar and the temperature values corresponding to all pixel points in the pseudo-color infrared image can be obtained. The minimum pixel value in the highest temperature zone can be obtained by obtaining the average value of the pixel values in the uppermost row in the highest temperature zone, and the maximum pixel value in the lowest temperature zone can be obtained by obtaining the average value of the pixel values in the lowermost row in the lowest temperature zone. average is obtained. In the above, the rectangular frame of the temperature data area is distributed vertically, and the temperature is distributed in a descending manner from top to bottom as an example.

For example, the maximum temperature is t ₁ , the minimum pixel value is l ₁ , the minimum temperature is t ₂ , the maximum pixel value is l ₂ , and the linear correlation coefficient is K, then by t ₁ =K*l ₁ +A, and t ₂ =K*l ₂ +A forms a system of equations, and A is a constant. The linear correlation coefficient K is obtained by solving the system of equations.

Further, the pseudo-color infrared image is an image of an infrared device detecting a power generation component. After the step S8 of estimating the temperature values corresponding to all the pixels in the pseudo-color infrared image according to the linear correlation coefficient, the method includes:

S81: Determine whether there is a specific pixel whose temperature value is greater than a preset threshold;

S82: If yes, obtain position information of the specific pixel in the pseudo-color infrared image;

S83: Determine the heating fault point of the power generation component according to the position information in the pseudo-color infrared image and the position mapping relationship of the power generation component detected by the infrared device.

In the embodiments of the present application, the above-mentioned power generation components include but are not limited to power equipment such as power transmission and transformation equipment. Taking the live detection of power transmission and transformation equipment in the power sector as an example, by monitoring the temperature data of each structural component in the power transmission and transformation equipment in time, the heating fault point can be found in time to ensure the safety of electricity consumption.

2 , a device for detecting temperature according to an embodiment of the present application includes:

The first acquisition module 1 is used to acquire the pseudo-color infrared image obtained in the infrared detection process;

Intercepting module 2, for intercepting the temperature data area in the pseudo-color infrared image;

Segmentation module 3, for segmenting the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

A comparison module 4 is used to compare the sub-picture with the templates in the preset template library, and respectively determine the specified template with the highest similarity with each of the sub-pictures;

The obtaining module 5 is used for obtaining the temperature range corresponding to the temperature data area according to the number of the specified template.

The explanation of the embodiments of the apparatus in the present application and the explanation of the corresponding parts of the applicable method will not be repeated.

Further, the temperature data area also includes a grayscale range, and the segmentation module 3 includes:

A determination submodule for determining the current grayscale threshold within the grayscale range corresponding to the temperature data area;

A partition sub-module for dividing the grayscale range of the temperature data region into two regions by the current grayscale threshold, including a first region smaller than the current grayscale threshold and a region greater than or equal to the current grayscale threshold the second area;

a calculation sub-module for calculating the grayscale variance of the first region and the second region under the current grayscale threshold;

an adjustment submodule, configured to dynamically adjust the grayscale threshold within the grayscale range corresponding to the temperature data region, and determine the maximum grayscale variance corresponding to the first region and the second region;

As a sub-module, the gray threshold corresponding to the maximum gray variance is used as the segmentation threshold;

A segmentation sub-module, configured to segment the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively by using the segmentation threshold.

Further, the calculation sub-module includes:

a first calculation unit, configured to calculate the first probability of each gray value distribution in the first area, and calculate the second probability of each gray value distribution in the second area;

A second calculation unit, configured to calculate a first average gray value corresponding to the first region according to each of the first probabilities, and calculate a second average gray value corresponding to the second region according to each of the second probabilities ;

a third calculation unit, configured to calculate the total average gray level of the temperature data region according to the first average gray value and the second average gray value;

a fourth calculation unit, configured to calculate the first average gray value according to the total average gray value, the first average gray value, the second average gray value, the first probability and the second probability The grayscale variance of the region and the second region under the current grayscale threshold.

Further, the sub-modules are divided, including:

a first determining unit, configured to determine a first boundary between the temperature data area and the background area, and a second boundary where the highest temperature number in the temperature data area is located, and the lowest temperature in the temperature data area according to the segmentation threshold The third frontier where the warm numbers are located;

A second determining unit, configured to determine the boundary of the sub-picture corresponding to the highest temperature area according to the first boundary and the second boundary, and determine the lowest temperature area according to the first boundary and the third boundary The border of the corresponding sub-picture;

The intercepting unit is configured to intercept the sub-picture corresponding to the highest temperature area according to the boundary of the sub-picture corresponding to the highest temperature area, and intercept the sub-picture corresponding to the lowest temperature area according to the boundary of the sub-picture corresponding to the lowest temperature area subimage.

Further, the second boundary and the third boundary are in the lateral direction of the temperature data area, and dividing submodules includes:

a judging unit for judging whether a specified pixel point on the second boundary is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary;

A search unit, configured to search for a gray level fusion point in a longitudinal direction passing through the designated pixel point and parallel to the temperature data area if it is not within the boundary threshold range of gray level change;

a replacement unit, configured to replace the gray-scale fusion point with the specified pixel point on the second boundary;

A correction unit, configured to correct all the pixels on the second boundary according to the correction method of the specified pixel point, and correct the third boundary according to the correction method of the second boundary.

Further, the device for detecting temperature includes:

a second acquisition module, configured to acquire the minimum pixel value within the highest temperature region and the maximum pixel value within the lowest temperature region;

a calculation module, configured to calculate the linear correlation coefficient between the pixel value and the temperature value according to the corresponding relationship between the maximum temperature and the minimum pixel value, and the corresponding relationship between the minimum temperature and the maximum pixel value;

The estimation module is used for estimating the temperature values corresponding to all the pixel points in the pseudo-color infrared image according to the linear correlation coefficient.

Further, the pseudo-color infrared image is an image of an infrared device detecting a power generation component, and a device for detecting temperature includes:

a judgment module for judging whether there is a specific pixel whose temperature value is greater than a preset threshold;

a third acquisition module, configured to acquire position information of the specific pixel in the pseudo-color infrared image if there is a specific pixel whose temperature value is greater than a preset threshold;

and a determination module, configured to determine the heating fault point of the power generation component according to the position information in the pseudo-color infrared image and the position mapping relationship of the power generation component detected by the infrared device.

Referring to FIG. 3 , an embodiment of the present application further provides a computer device. The computer device may be a server, and its internal structure may be as shown in FIG. 3 . The computer device includes a processor, memory, a network interface, and a database connected by a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store all the data required for the process of detecting the temperature. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of detecting temperature.

The above-mentioned method for detecting temperature by the processor includes: acquiring a pseudo-color infrared image obtained in an infrared detection process; intercepting a temperature data area in the pseudo-color infrared image; dividing the temperature data area according to a preset segmentation method The sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively, wherein the sub-picture includes only one number; the sub-picture is compared with the templates in the preset template library, and the corresponding sub-pictures are determined respectively. The specified template with the highest similarity; according to the number of the specified template, the temperature range corresponding to the temperature data area is obtained.

The above-mentioned computer equipment, based on the position of the temperature bar in the pseudo-color infrared image, is relatively fixed, and the identification fonts of the highest temperature area and the lowest temperature area above and below the temperature bar are also consistent, by intercepting the temperature data area from the temperature bar. The sub-images of the highest temperature area and the lowest temperature area, and using a simple and efficient digital template matching method, the temperature range in the temperature bar in the pseudo-color infrared image can be obtained, with a small workload and high recognition efficiency.

In one embodiment, the temperature data area further includes a grayscale range, and the step of dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area by the processor from the temperature data area according to a preset segmentation method includes the following steps: : determine the current grayscale threshold within the grayscale range corresponding to the temperature data area; divide the grayscale range of the temperature data area into two areas by the current grayscale threshold, including the grayscale range smaller than the current grayscale threshold The first area and the second area greater than or equal to the current grayscale threshold; calculate the grayscale variance of the first area and the second area under the current grayscale threshold; in the temperature data area corresponding to Dynamically adjust the grayscale threshold within the grayscale range to determine the maximum grayscale variance corresponding to the first area and the second area; use the grayscale threshold corresponding to the maximum grayscale variance as the segmentation threshold; pass the segmentation threshold Sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented.

In one embodiment, the step of calculating the grayscale variance of the first region and the second region under the current grayscale threshold by the processor includes: calculating the distribution of each grayscale value in the first region Calculate the first probability of each gray value distribution in the second area; calculate the first average gray value corresponding to the first area according to each of the first probability, and calculate the first average gray value corresponding to the first area according to each of the second Probabilistically calculate the second average gray value corresponding to the second area; calculate the total average gray value of the temperature data area according to the first average gray value and the second average gray value; The total average grayscale, the first average grayscale value, the second average grayscale value, the first probability, and the second probability, calculating the first area and the second area in the The grayscale variance at the current grayscale threshold.

In an embodiment, the step of dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area by the above-mentioned processor by using the dividing threshold includes: determining a first difference between the temperature data area and the background area according to the dividing threshold boundary, and a second boundary where the highest temperature number in the temperature data area is located, and a third boundary where the lowest temperature number is located in the temperature data area; determining the first boundary and the second boundary The boundary of the sub-picture corresponding to the highest temperature area is determined according to the first boundary and the third boundary of the sub-picture corresponding to the lowest temperature area; according to the boundary of the sub-picture corresponding to the highest temperature area, intercept the The sub-picture corresponding to the highest temperature area is intercepted according to the boundary of the sub-picture corresponding to the lowest temperature area.

In one embodiment, the second boundary and the third boundary are in the lateral direction of the temperature data area, and the processor determines the first boundary between the temperature data area and the background area according to the segmentation threshold, and after the step of the second boundary where the highest temperature number is located in the temperature data area and the third boundary where the lowest temperature number is located in the temperature data area, including: judging whether the specified pixel point on the second boundary is It is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary; In the longitudinal direction, search for gray-scale fusion points; replace the gray-scale fusion points on the specified pixel points on the second boundary; modify all the pixel points on the second boundary according to the modification method of the specified pixel points , and modify the third border according to the modification method of the second border.

In one embodiment, after the above-mentioned processor obtains the temperature range corresponding to the temperature data area according to the number of the specified template, the step includes: obtaining the minimum pixel value in the highest temperature area and the lowest temperature The maximum pixel value in the area; according to the corresponding relationship between the maximum temperature and the minimum pixel value, and the corresponding relationship between the minimum temperature and the maximum pixel value, calculate the linear correlation coefficient between the pixel value and the temperature value; according to the linear correlation coefficient The temperature values corresponding to all the pixel points in the pseudo-color infrared image are estimated respectively.

In one embodiment, the pseudo-color infrared image is an image of an infrared device detecting a power-generating component, and after the step of estimating the temperature values corresponding to all pixels in the pseudo-color infrared image by the processor according to the linear correlation coefficient, the method includes: : determine whether there is a specific pixel with a temperature value greater than a preset threshold; if so, obtain the position information of the specific pixel in the pseudo-color infrared image; according to the position information in the pseudo-color infrared image, and the infrared The device detects the positional mapping relationship of the power generation components, and determines the heating failure points of the power generation components.

Those skilled in the art can understand that the structure shown in FIG. 3 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied.

An embodiment of the present application further provides a computer-readable storage medium, the computer-readable storage medium may be non-volatile or volatile, and a computer program is stored thereon, and the computer program is implemented when executed by a processor A method for detecting temperature, comprising: acquiring a pseudo-color infrared image obtained in an infrared detection process; intercepting a temperature data area in the pseudo-color infrared image; dividing the highest temperature area and the lowest temperature area from the temperature data area according to a preset segmentation method The sub-pictures corresponding to the temperature zones respectively, wherein the sub-pictures only include a number; the sub-pictures are compared with the templates in the preset template library, and the designations with the highest similarity with each of the sub-pictures are determined respectively. template; obtain the temperature range corresponding to the temperature data area according to the number of the specified template.

The above-mentioned computer-readable storage medium is based on the fact that the position of the temperature bar in the pseudo-color infrared image is relatively fixed, and the identification fonts of the highest temperature area and the lowest temperature area above and below the temperature bar are also consistent. , cut out the sub-pictures of the highest temperature area and the lowest temperature area, and use a simple and efficient digital template matching method to obtain the temperature range in the temperature bar in the pseudo-color infrared image, with small workload and high recognition efficiency.

Those of ordinary skill in the art can understand that all or part of the process in the method of the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium , when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other medium provided in this application and used in the embodiments may include non-volatile and/or volatile memory. Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, device, article or method comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, apparatus, article or method. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, apparatus, article, or method that includes the element.

The above are only the preferred embodiments of the present application, and are not intended to limit the scope of the patent of the present application. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present application, or directly or indirectly applied to other related The technical field is similarly included in the scope of patent protection of this application.

Claims

A method of detecting temperature, comprising:

Obtain the pseudo-color infrared image obtained during the infrared detection process;

intercepting the temperature data area in the pseudo-color infrared image;

The sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

Comparing the sub-pictures with the templates in the preset template library, respectively determining the specified template with the highest similarity with each of the sub-pictures;

According to the number of the specified template, the temperature range corresponding to the temperature data area is obtained.
The method for detecting temperature according to claim 1, wherein the temperature data area further includes a gray scale range, and the highest temperature area and the lowest temperature area are divided from the temperature data area according to a preset segmentation method respectively corresponding to The steps for the subimage include:

Determine the current grayscale threshold within the grayscale range corresponding to the temperature data area;

Dividing the grayscale range of the temperature data region into two regions by using the current grayscale threshold, including a first region smaller than the current grayscale threshold and a second region greater than or equal to the current grayscale threshold;

Calculate the grayscale variance of the first area and the second area under the current grayscale threshold;

Dynamically adjust the grayscale threshold within the grayscale range corresponding to the temperature data region, and determine the maximum grayscale variance corresponding to the first region and the second region;

Using the grayscale threshold corresponding to the maximum grayscale variance as the segmentation threshold;

The sub-pictures respectively corresponding to the highest temperature area and the lowest temperature area are segmented by the segmentation threshold.
The method for detecting temperature according to claim 2, wherein the step of calculating the grayscale variance of the first region and the second region under the current grayscale threshold comprises:

calculating a first probability of each gray value distribution in the first area, and calculating a second probability of each gray value distribution in the second area;

Calculate a first average grayscale value corresponding to the first region according to each of the first probabilities, and calculate a second average grayscale value corresponding to the second region according to each of the second probabilities;

calculating the total average gray level of the temperature data region according to the first average gray value and the second average gray value;

Calculate the first area and the second area according to the total average grayscale, the first average grayscale value, the second average grayscale value, the first probability, and the second probability The grayscale variance under the current grayscale threshold.
The method for detecting temperature according to claim 2, wherein the step of dividing the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively by the segmentation threshold comprises:

The first boundary between the temperature data area and the background area, the second boundary where the highest temperature number in the temperature data area is located, and the third boundary where the lowest temperature number in the temperature data area is located are determined according to the segmentation threshold. boundary;

Determine the boundary of the sub-picture corresponding to the highest temperature area according to the first boundary and the second boundary, and determine the boundary of the sub-picture corresponding to the lowest temperature area according to the first boundary and the third boundary;

The sub-picture corresponding to the highest temperature region is intercepted according to the boundary of the sub-picture corresponding to the highest temperature region, and the sub-picture corresponding to the lowest temperature region is intercepted according to the boundary of the sub-picture corresponding to the lowest temperature region.
The method for detecting temperature according to claim 4, wherein the second boundary and the third boundary are in a lateral direction of the temperature data region, and the temperature data region is determined according to the segmentation threshold to be the same as the temperature data region. After the steps of the first boundary of the background area, the second boundary where the highest temperature number is located in the temperature data area, and the third boundary where the lowest temperature number is located in the temperature data area, the steps include:

Determine whether the specified pixel point on the second boundary is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary;

If not, searching for a gray-scale fusion point in the longitudinal direction passing through the designated pixel point and parallel to the temperature data area;

replacing the gray-scale fusion point with the specified pixel point on the second boundary;

All pixels on the second boundary are corrected according to the correction method of the specified pixel point, and the third boundary is corrected according to the correction method of the second boundary.
The method for detecting temperature according to claim 1, wherein after the step of obtaining the temperature range corresponding to the temperature data area according to the number of the specified template, the method comprises:

obtaining the minimum pixel value within the highest temperature region, and the maximum pixel value within the lowest temperature region;

According to the corresponding relationship between the maximum temperature and the minimum pixel value, and the corresponding relationship between the minimum temperature and the maximum pixel value, calculate the linear correlation coefficient between the pixel value and the temperature value;

The temperature values corresponding to all pixel points in the pseudo-color infrared image are estimated according to the linear correlation coefficient.
The method for detecting temperature according to claim 6, wherein the pseudo-color infrared image is an image of an infrared device detecting a power-generating component, and all the pixels in the pseudo-color infrared image are estimated according to the linear correlation coefficient, respectively. After the steps corresponding to the temperature values, include:

Determine whether there is a specific pixel whose temperature value is greater than a preset threshold;

If so, obtain the position information of the specific pixel in the pseudo-color infrared image;

According to the position information in the pseudo-color infrared image and the position mapping relationship of the power generation components detected by the infrared device, the heating failure point of the power generation components is determined.
A device for detecting temperature, comprising:

a first acquisition module, used for acquiring a pseudo-color infrared image obtained during the infrared detection process;

an interception module, used for intercepting the temperature data area in the pseudo-color infrared image;

a segmentation module, configured to segment the sub-pictures corresponding to the highest temperature area and the lowest temperature area respectively from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

a comparison module, configured to compare the sub-pictures with the templates in the preset template library, and respectively determine the specified template with the highest similarity with each of the sub-pictures;

The obtaining module is used to obtain the temperature range corresponding to the temperature data area according to the number of the specified template.
A computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein the processor implements a method for detecting temperature when the processor executes the computer program, the method comprising:

Obtain the pseudo-color infrared image obtained during the infrared detection process;

intercepting the temperature data area in the pseudo-color infrared image;

The sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

Comparing the sub-pictures with the templates in the preset template library, respectively determining the specified template with the highest similarity with each of the sub-pictures;

According to the number of the specified template, the temperature range corresponding to the temperature data area is obtained.
The computer device according to claim 9, wherein the temperature data area further includes a gray scale range, and the subsections corresponding to the highest temperature area and the lowest temperature area are divided from the temperature data area according to a preset segmentation method. Steps for pictures, including:

Determine the current grayscale threshold within the grayscale range corresponding to the temperature data area;

Dividing the grayscale range of the temperature data region into two regions by using the current grayscale threshold, including a first region smaller than the current grayscale threshold and a second region greater than or equal to the current grayscale threshold;

Calculate the grayscale variance of the first area and the second area under the current grayscale threshold;

Dynamically adjust the grayscale threshold within the grayscale range corresponding to the temperature data region, and determine the maximum grayscale variance corresponding to the first region and the second region;

Using the grayscale threshold corresponding to the maximum grayscale variance as the segmentation threshold;

The sub-pictures respectively corresponding to the highest temperature area and the lowest temperature area are segmented by the segmentation threshold.
The computer device according to claim 10, wherein the step of calculating the grayscale variance of the first region and the second region under the current grayscale threshold comprises:

calculating a first probability of each gray value distribution in the first area, and calculating a second probability of each gray value distribution in the second area;

Calculate a first average grayscale value corresponding to the first region according to each of the first probabilities, and calculate a second average grayscale value corresponding to the second region according to each of the second probabilities;

calculating the total average gray level of the temperature data region according to the first average gray value and the second average gray value;

Calculate the first area and the second area according to the total average grayscale, the first average grayscale value, the second average grayscale value, the first probability, and the second probability The grayscale variance under the current grayscale threshold.
The computer device according to claim 10, wherein the step of dividing the sub-pictures corresponding to the highest temperature region and the lowest temperature region respectively by the segmentation threshold comprises:

The first boundary between the temperature data area and the background area, the second boundary where the highest temperature number in the temperature data area is located, and the third boundary where the lowest temperature number in the temperature data area is located are determined according to the segmentation threshold. boundary;

Determine the boundary of the sub-picture corresponding to the highest temperature area according to the first boundary and the second boundary, and determine the boundary of the sub-picture corresponding to the lowest temperature area according to the first boundary and the third boundary;

The sub-picture corresponding to the highest temperature region is intercepted according to the boundary of the sub-picture corresponding to the highest temperature region, and the sub-picture corresponding to the lowest temperature region is intercepted according to the boundary of the sub-picture corresponding to the lowest temperature region.
13. The computer device of claim 12, wherein the second boundary and the third boundary are in a lateral direction of the temperature data area, and the temperature data area and the background area are determined according to the segmentation threshold After the steps of the first boundary of the temperature data area, the second boundary where the highest temperature number is located in the temperature data area, and the third boundary where the lowest temperature number is located in the temperature data area, the steps include:

Determine whether the specified pixel point on the second boundary is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary;

If not, searching for a gray-scale fusion point in the longitudinal direction passing through the designated pixel point and parallel to the temperature data area;

replacing the gray-scale fusion point with the specified pixel point on the second boundary;

All pixels on the second boundary are corrected according to the correction method of the specified pixel point, and the third boundary is corrected according to the correction method of the second boundary.
The computer device according to claim 9, wherein, after the step of obtaining the temperature range corresponding to the temperature data area according to the numbers of the specified template, the step includes:

obtaining the minimum pixel value within the highest temperature region, and the maximum pixel value within the lowest temperature region;

According to the corresponding relationship between the maximum temperature and the minimum pixel value, and the corresponding relationship between the minimum temperature and the maximum pixel value, calculate the linear correlation coefficient between the pixel value and the temperature value;

The temperature values corresponding to all pixel points in the pseudo-color infrared image are estimated according to the linear correlation coefficient.
The computer device according to claim 14, wherein the pseudo-color infrared image is an image of an infrared device detecting a power-generating component, and estimating the corresponding corresponding pixels of all pixels in the pseudo-color infrared image according to the linear correlation coefficient After the temperature value steps, include:

Determine whether there is a specific pixel whose temperature value is greater than a preset threshold;

If so, obtain the position information of the specific pixel in the pseudo-color infrared image;

According to the position information in the pseudo-color infrared image and the position mapping relationship of the power generation components detected by the infrared device, the heating failure point of the power generation components is determined.
A computer-readable storage medium on which a computer program is stored, wherein, when the computer program is executed by a processor, a method for detecting temperature, the method comprising:

Obtain the pseudo-color infrared image obtained during the infrared detection process;

intercepting the temperature data area in the pseudo-color infrared image;

The sub-pictures corresponding to the highest temperature area and the lowest temperature area are segmented from the temperature data area according to a preset segmentation method, wherein the sub-picture only includes a number;

Comparing the sub-pictures with the templates in the preset template library, respectively determining the specified template with the highest similarity with each of the sub-pictures;

According to the number of the specified template, the temperature range corresponding to the temperature data area is obtained.
The computer-readable storage medium according to claim 16, wherein the temperature data area further includes a gray scale range, and the highest temperature area and the lowest temperature area are divided from the temperature data area according to a preset segmentation method, respectively. The steps of the corresponding sub-picture include:

Determine the current grayscale threshold within the grayscale range corresponding to the temperature data area;

Dividing the grayscale range of the temperature data region into two regions by using the current grayscale threshold, including a first region smaller than the current grayscale threshold and a second region greater than or equal to the current grayscale threshold;

Calculate the grayscale variance of the first area and the second area under the current grayscale threshold;

Dynamically adjust the grayscale threshold within the grayscale range corresponding to the temperature data region, and determine the maximum grayscale variance corresponding to the first region and the second region;

Using the grayscale threshold corresponding to the maximum grayscale variance as the segmentation threshold;

The sub-pictures respectively corresponding to the highest temperature area and the lowest temperature area are segmented by the segmentation threshold.
The computer-readable storage medium according to claim 17, wherein the step of calculating the grayscale variance of the first region and the second region under the current grayscale threshold comprises:

calculating a first probability of each gray value distribution in the first area, and calculating a second probability of each gray value distribution in the second area;

Calculate a first average grayscale value corresponding to the first region according to each of the first probabilities, and calculate a second average grayscale value corresponding to the second region according to each of the second probabilities;

calculating the total average gray level of the temperature data region according to the first average gray value and the second average gray value;

Calculate the first area and the second area according to the total average grayscale, the first average grayscale value, the second average grayscale value, the first probability, and the second probability The grayscale variance under the current grayscale threshold.
The computer-readable storage medium according to claim 17, wherein the step of dividing the sub-pictures corresponding to the highest temperature region and the lowest temperature region respectively by the segmentation threshold comprises:

A first boundary between the temperature data area and the background area, a second boundary where the highest temperature number in the temperature data area is located, and a third boundary where the lowest temperature number in the temperature data area is located are determined according to the segmentation threshold boundary;

Determine the boundary of the sub-picture corresponding to the highest temperature area according to the first boundary and the second boundary, and determine the boundary of the sub-picture corresponding to the lowest temperature area according to the first boundary and the third boundary;

The sub-picture corresponding to the highest temperature region is intercepted according to the boundary of the sub-picture corresponding to the highest temperature region, and the sub-picture corresponding to the lowest temperature region is intercepted according to the boundary of the sub-picture corresponding to the lowest temperature region.
19. The computer-readable storage medium of claim 19, wherein the second boundary and the third boundary are in a lateral direction of the temperature data region, the temperature data region being determined according to the segmentation threshold After the step with the first boundary of the background area, the second boundary where the highest temperature number is located in the temperature data area, and the third boundary where the lowest temperature number is located in the temperature data area, the steps include:

Determine whether the specified pixel point on the second boundary is within the boundary threshold range of grayscale change, wherein the specified pixel point is any pixel point on the second boundary;

If not, searching for a gray-scale fusion point in the longitudinal direction passing through the designated pixel point and parallel to the temperature data area;

replacing the gray-scale fusion point with the specified pixel point on the second boundary;

All pixels on the second boundary are corrected according to the correction method of the specified pixel point, and the third boundary is corrected according to the correction method of the second boundary.