Body temperature anomaly detection method based on infrared thermal imaging

Abstract

The invention relates to the technical field of image data processing, in particular to an infrared thermal imaging-based body temperature anomaly detection method, which comprises the following steps of: firstly, performing enhancement processing on an initial image according to different block sizes by using a CLAHE algorithm; then, calculating the direction consistency degree of each pixel point in the enhanced image under each block size, and quantifying the noise confidence weight of each pixel point through the information entropy of the pixel points in the neighbor range; then, calculating an average value of products of noise confidence weights and direction consistency degrees of all pixel points in the same enhanced image to obtain a feasible degree of the size of each block; and finally, according to a weight coefficient of feasible degree conversion, carrying out weighted fusion on gray values of the same pixel position in the enhanced images under all block sizes so as to realize abnormal temperature detection based on the generated target enhanced image. According to the invention, the image enhancement process can be optimized, and the abnormal temperature of the livestock and poultry body surface can be accurately detected.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing. More specifically, the present invention relates to a method for detecting abnormal body temperature based on infrared thermal imaging. Background Art

[0002] In large-scale livestock and poultry farming, body temperature is a key indicator reflecting the health status of animals. Traditional contact thermometry methods are prone to causing stress responses in animals and are difficult to achieve real-time monitoring of groups. Infrared thermal imaging technology, with its non-contact and high-efficiency advantages, provides an important means for detecting abnormal body temperature in livestock and poultry. Through the temperature distribution characteristics in infrared images, suspected diseased individuals can be quickly located, realizing early warning of diseases and precise prevention and control, which is of great significance for ensuring the economic benefits of the breeding industry and animal welfare.

[0003] However, in the detection of abnormal body temperature in livestock and poultry, although infrared thermal imaging technology can achieve non-contact and large-scale body temperature monitoring, the original infrared images often have problems such as low contrast, blurred details, and noise interference due to environmental noise, insufficient equipment resolution, and the complexity of body surface thermal radiation. This makes it difficult to identify subtle temperature abnormalities (such as local inflammation or early infection). Therefore, image enhancement of the collected original infrared images is a key preprocessing step that can improve the detection accuracy.

[0004] Currently, infrared image enhancement mostly uses the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, and usually uses a fixed block size to enhance the image. However, when the block size is too large, the algorithm is difficult to effectively enhance the contrast of local tiny temperature abnormal areas in the image, resulting in insufficient image enhancement; while when the block size is too small, although it can improve local details, it is easy to generate artifact noise due to excessive correction of local histograms, interfering with the true temperature distribution, and unable to balance the requirements of maintaining the global temperature distribution characteristics and enhancing local details, resulting in poor image enhancement effects and affecting subsequent analysis results. Summary of the Invention

[0005] In order to solve the problem that when using the conventional CLAHE algorithm to enhance infrared images, it is impossible to balance the requirements of maintaining the global temperature distribution characteristics and enhancing local details, resulting in poor image enhancement effects and then affecting the accuracy of subsequent abnormal temperature detection, the present invention provides a method for detecting abnormal body temperature based on infrared thermal imaging. The method includes: Obtain the temperature distribution map of the livestock and poultry body surface and perform preprocessing to obtain an initial image; Based on different block sizes, use the CLAHE algorithm to perform image enhancement on the initial image respectively, and calculate the direction consistency of each pixel point in the enhanced image under each block size. The direction consistency represents the consistency between the gradient direction of the pixel point and the gradient direction of the corresponding pixel point in different enhanced images; Quantify the noise confidence weight of each pixel in the enhanced image under each block size through the information entropy of all pixels within the neighborhood of the pixel, and calculate the average value of the product of the noise confidence weights and the direction consistency of all pixels in the same enhanced image, so as to obtain the feasibility of each block size; Convert the feasibility into weight coefficients with a sum of 1, and according to the weight coefficients, perform weighted summation on the gray values of the same pixel position in the enhanced images under all block sizes to obtain the target enhanced image, so as to detect the abnormal temperature on the body surface of livestock and poultry based on the target enhanced image.

[0006] In the present invention, first, based on different block sizes, the initial image is enhanced by the CLAHE algorithm respectively, and then the direction consistency of each pixel in the enhanced image under each block size is calculated. By using the feature that the local structures of the pixels in the enhanced image under the appropriate block size have a high degree of consistency, the appropriateness of each block size is evaluated. After that, the noise confidence weight is quantified, which can reduce the influence of noise on the evaluation of the appropriateness of each block size. Therefore, the feasibility of each block size can be accurately measured through the calculated feasibility. Then, by converting the feasibility into weight coefficients, the gray values of the same pixel position in the enhanced images under all block sizes are fused, avoiding the problems of insufficient enhancement or artifacts caused by improper selection of the block size, optimizing the image enhancement process, and providing an accurate image basis for the subsequent detection of abnormal temperature on the body surface of livestock and poultry.

[0007] Preferably, calculating the direction consistency of each pixel in the enhanced image under each block size includes: For the enhanced image under each block size, calculate the direction consistency of any pixel according to the difference between the gradient direction angle of any pixel in any enhanced image and the gradient direction angles of the corresponding pixels in all the remaining enhanced images. The direction consistency is negatively correlated with the difference.

[0008] The present invention calculates the direction consistency of each pixel by using the local structure of the pixels in the enhanced image under the appropriate block size, especially the feature that the gradient directions have a high degree of consistency, and can evaluate the appropriateness of enhancing the initial image with each block size.

[0009] Preferably, the difference is a comprehensive difference, and the comprehensive difference satisfies the following relational expression: ; In the formula, is the comprehensive difference between the gradient direction angle of the th pixel in the enhanced image under the th block size and the gradient direction angle of the th pixel in the enhanced image under the th block size; , are the sizes of the -th and -th block sizes respectively; is the size of the maximum block size; is the ordinal number of the maximum block size; , are the -th and -th block sizes respectively, and the gradient direction angle of the -th pixel point in the enhanced image under the is the absolute value symbol; is a function whose return value is the minimum value; is the natural exponential function.

[0010] When calculating the difference in the gradient direction angles of the corresponding pixel points in the enhanced images under different block sizes, the present invention takes into account the influence of the difference in the sizes of the block sizes, and can differentially consider the image enhancement results under different block sizes.

[0011] Preferably, when calculating the direction consistency of any pixel point, the following relational expression is satisfied: ; In the formula, is the direction consistency of the -th pixel point in the enhanced image under the -th block size; is the comprehensive difference between the gradient direction angle of the -th pixel point and the gradient direction angle of the -th pixel point in the enhanced image under the -th block size and the -th pixel point in the enhanced image under the [[ID=�5]] is the ordinal number of the maximum block size; is the natural exponential function.

[0012] Preferably, the method for obtaining the gradient direction angle includes: Using the Sobel operator to obtain the horizontal gradient value and the vertical gradient value of any pixel point, and calculating the arctangent value of the ratio of the vertical gradient value to the horizontal gradient value to obtain the gradient direction angle of any pixel point.

[0013] Preferably, the size of the maximum block size is the largest positive integer less than half of the width of the initial image, and the difference in the sizes of the block sizes with adjacent ordinal numbers is a preset integral multiple of 2.

[0014] By restricting the maximum block size, the present invention ensures that the initial image can be divided into at least two or more image blocks.

[0015] Preferably, the noise confidence weights of each pixel in the enhanced image under each block size are quantified by the information entropy of all pixels within the neighborhood of the pixel, including: For each pixel in the enhanced image under each block size, the information entropy of all pixels within the neighborhood of any pixel is used as the local entropy of the any pixel, and the difference between 1 and the local entropy is used as the noise confidence weight of the any pixel.

[0016] Preferably, the feasibility of each block size satisfies the following relational expression: ; In the formula, is the feasibility of the th block size; is the local entropy of the th pixel in the enhanced image under the th block size; is the direction consistency of the th pixel in the enhanced image under the th block size; is the th block size;

[0017] Preferably, based on the target enhanced image, detecting the abnormal temperature on the body surface of livestock and poultry includes: Obtaining a preset gray threshold. If the gray value of any pixel in the target enhanced image is less than the gray threshold, it is determined that the area where the any pixel is located is a normal temperature area; If it is greater than or equal to the gray threshold, the area where the any pixel is located is marked as an abnormal temperature area.

[0018] Preferably, the method for obtaining the initial image includes: The temperature distribution map is sequentially subjected to graying processing and filtering processing, and the initial image is obtained after the filtering is completed.

[0019] Through a series of preprocessing operations, the present invention can provide a clearer and more accurate initial image basis for subsequent image enhancement.

[0020] The present invention has the following effects: Through CLAHE image enhancement based on different block sizes and combined with the comprehensive evaluation of direction consistency and noise confidence weight, the present invention can effectively take into account the global temperature distribution characteristics of the image and the local detail enhancement requirements, realize the adaptive fusion of enhanced images with different block sizes, avoid the problems of insufficient enhancement or artifacts caused by a single fixed block size, optimize the image enhancement process, and enable the accurate detection of abnormal body temperatures on the surfaces of livestock and poultry based on the obtained target enhanced image. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic flow chart of the steps of the method for detecting abnormal body temperature based on infrared thermal imaging according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The following will describe in detail the specific embodiments of the present invention with reference to the accompanying drawings.

[0023] Refer to Figure 1 , the method for detecting abnormal body temperature based on infrared thermal imaging includes steps S1 - S4, which are specifically as follows: S1: Obtain the temperature distribution map of the surface of livestock and poultry and perform pre - processing to obtain an initial image.

[0024] It should be noted that the temperature distribution on the surface of healthy livestock and poultry follows specific rules and ranges, and when livestock and poultry have diseases or physiological abnormalities, it often leads to changes in the surface temperature, such as local fever or overall abnormal body temperature, etc. Therefore, the real - time monitoring of the body temperature of livestock and poultry can be achieved through the temperature distribution on the surface of livestock and poultry.

[0025] Specifically, an infrared thermal imager or an infrared camera can be used to collect the infrared thermal imaging image of livestock and poultry to obtain a temperature distribution map that can intuitively reflect the temperature distribution on the surface of livestock and poultry. It should be noted that since the captured image may be a pseudo - color image and may contain noise, a series of pre - processing operations need to be performed on the collected temperature distribution map to improve the image quality.

[0026] In an exemplary embodiment of the present invention, the determination of the initial image can be achieved through the following steps: Perform grayscale processing and filtering on the temperature distribution map in sequence, and obtain the initial image after filtering is completed.

[0027] Optionally, algorithms such as Gaussian filtering, median filtering, or mean filtering can be used to perform filtering on the collected temperature distribution map of the surface of livestock and poultry. This embodiment does not make a special limitation on the selected filtering method.

[0028] It should be noted that the processes of grayscale processing and filtering of the image are prior arts, and this embodiment will not elaborate on them here.

[0029] S2: Based on different block sizes, the initial image is enhanced using the CLAHE algorithm. The directional consistency of each pixel in the enhanced image at each block size is calculated. The directional consistency represents the consistency of the gradient direction of the pixel with the gradient direction of the corresponding pixel in different enhanced images.

[0030] It should be noted that when using the conventional CLAHE (Contrast Limited Adaptive Histogram Equalization) algorithm for image enhancement, a fixed-size block size is usually used to obtain an enhanced image. However, when the block size is too large, the image area contained in each block becomes larger, and the local features are diluted, which will result in the inability to fully enhance the image; when the selected block size is too small, artifacts may be introduced. Therefore, in order to fully enhance the image while avoiding too many artifacts as much as possible, the present invention sets a plurality of different block sizes, and then evaluates the similarity of the local structure of the enhanced image at each block size with the enhanced image at other block sizes through the directional consistency of the pixel points, thereby providing a data basis for subsequent calculations.

[0031] In an exemplary embodiment of the present invention, the size of the maximum block size is a maximum positive integer smaller than half of the width of the initial image, and the difference between the sizes of ordinal adjacent block sizes is a preset integer multiple of 2.

[0032] The width of the initial image refers to the number of pixels contained in the shorter side of the initial image.

[0033] Specifically, you can set the block size set , satisfying the relationship: , where For the The size of each block; is the ordinal number of the block size; The ordinal number of the maximum block size.

[0034] For example, when When , the set block sizes include 8×8, 16×16, 32×32 and 64×64, so that the initial image can be enhanced respectively based on the above four block sizes using the CLAHE algorithm to obtain enhanced images under the four block sizes.

[0035] The purpose of limiting the maximum block size is to ensure that the initial image can be divided into at least two blocks, and the purpose of limiting the difference in the sizes of ordinal adjacent blocks to an integer multiple of 2 is to allow the block sizes to cover a range from small to large, thereby adapting to abnormal body temperature areas of different scales in the initial image.

[0036] It should be noted that during the process of respectively enhancing the initial image using the CLAHE algorithm based on different block sizes, the contrast limiting factor for each set block size is 2.0. This value is a conventional value selected in the CLAHE algorithm and is not specifically limited in this embodiment. Among them, in the case of a known block size, the process of image enhancement using the CLAHE algorithm is a prior art and will not be described in detail in this embodiment.

[0037] In an exemplary embodiment of the present invention, the determination of the direction consistency of each pixel point in the enhanced image for each block size can be achieved through the following steps: For the enhanced image for each block size, according to the difference between the gradient direction angle of any pixel point in any enhanced image and the gradient direction angles of the corresponding pixel points in all the remaining enhanced images, calculate the direction consistency of any pixel point. The direction consistency is negatively correlated with the difference.

[0038] It should be noted that if the block size is appropriate, in the image enhanced at this size, the local structure of the pixel points (especially the gradient direction) has a high consistency with the results of other appropriate sizes. On the contrary, an inappropriate block size will lead to insufficient image enhancement or introduce artifacts, showing a significant difference from the CLAHE results at an appropriate block size. Therefore, based on this feature, the present invention can measure whether each block size is appropriate by evaluating the direction consistency of each pixel point.

[0039] In an exemplary embodiment of the present invention, the determination of the gradient direction angle of a pixel point can be achieved through the following steps: Use the Sobel operator to obtain the horizontal gradient value and the vertical gradient value of any pixel point, and calculate the arctangent value of the ratio of the vertical gradient value to the horizontal gradient value to obtain the gradient direction angle of any pixel point.

[0040] Optionally, the Prewitt operator, Roberts operator, etc. can also be used to obtain the horizontal gradient value and the vertical gradient value of the pixel point. The type of operator selected in this embodiment is not specifically limited. It should be noted that the process of using the Sobel operator to obtain the horizontal gradient value and the vertical gradient value is a prior art and will not be described in detail in this embodiment.

[0041] In an exemplary embodiment of the present invention, the difference between the gradient direction angles of the pixel points at the same pixel position in the enhanced images under different block sizes is a comprehensive difference, and this comprehensive difference satisfies the following relational expression: ; In the formula, is the th enhanced image under the The comprehensive difference between the gradient direction angle of the th pixel point and the gradient direction angle of the th pixel point in the enhanced image under the -th block size; are respectively the th block size and the th block size; is the size of the maximum block size; is the ordinal number of the maximum block size; 、 are respectively the th block size and the th block size of the gradient direction angle of the th pixel point in the enhanced image; is the absolute value symbol; is a function that returns the minimum value; is the natural exponential function; is 180 degrees.

[0042] Among them, reflects the size difference weight between the th block size and the th block size. When the difference between these two block sizes (the absolute value of the numerical difference) is small, if both of these two block sizes are appropriate block sizes, then the local features of the pixel points in the enhanced images under these two block sizes have high similarity. At this time, setting a larger size difference weight can differentially consider the image enhancement results under different block sizes.

[0043] reflects the difference in the gradient direction angle of the th pixel point in the enhanced images under these two block sizes. The smaller this value is, the more consistent the gradient direction of the th pixel point in the enhanced image under the th block size is with the gradient direction of the th pixel point in the enhanced image under the th block size, and the corresponding th pixel point in the enhanced image under the th block size has a relatively high direction consistency.

[0044] In another embodiment, the relationship formula: ; can also be directly used to evaluate the difference in the gradient direction angle of the pixel points at the same pixel position in the enhanced images under different block sizes.

[0045] Further, after determining the difference between the gradient direction angle of any pixel in the enhanced image under any block size and the gradient direction angles of the corresponding pixels in the enhanced images under other block sizes, the direction consistency of any pixel can be calculated based on all the obtained differences.

[0046] Specifically, the direction consistency of any pixel satisfies the following relational expression: ; In the formula, is the direction consistency of the th pixel in the enhanced image under the th block size; is the th block size, and is the comprehensive difference between the gradient direction angle of the th pixel in the enhanced image and the gradient direction angle of the th pixel in the enhanced image under the th block size; is the ordinal number of the maximum block size; is the natural exponential function.

[0047] In another embodiment, the negative correlation between the difference and the direction consistency can also be characterized by using the Sigmoid function.

[0048] Furthermore, the direction consistency calculation formula can be used to calculate the direction consistency of all pixels in the enhanced image under each block size, thereby providing a data basis for subsequent analysis processes.

[0049] S3: Quantify the noise confidence weight of each pixel in the enhanced image under each block size through the information entropy of all pixels within the neighborhood range of the pixel, and calculate the average value of the product of the noise confidence weights and the direction consistency of all pixels in the same enhanced image, to obtain the feasibility of each block size.

[0050] It should be noted that although the image has been filtered in the preprocessing stage, there may still be high-noise regions in the initial image. Therefore, in order to reduce the interference of noise on the evaluation of the image enhancement effect, the present invention further quantifies the noise confidence weight of each pixel, so as to accurately evaluate whether each block size is appropriate, that is, the feasibility of each block size.

[0051] In an exemplary embodiment of the present invention, the determination of the noise confidence weight of each pixel can be achieved through the following steps: For each pixel in the enhanced image under each block size, the information entropy of all pixels within the neighborhood of any pixel is used as the local entropy of that pixel, and the difference between 1 and the local entropy is used as the noise confidence weight of that pixel.

[0052] Among them, the information entropy reflects the complexity and uncertainty index of the gray value distribution within the neighborhood of any pixel. The larger this value, the more chaotic the gray value distribution around that pixel, indicating that there may be noise around that pixel, and the corresponding noise confidence weight of that pixel is relatively low.

[0053] The neighborhood range refers to the local range centered on any pixel with a neighborhood size of an empirical value (such as 3×3). Thus, based on the information entropy of the gray values of all pixels within the neighborhood of each pixel, the noise confidence weights of all pixels in the enhanced image under each block size can be determined.

[0054] Furthermore, for the enhanced image under each block size, the average value of the product of the noise confidence weights and the direction consistency of all pixels in the same enhanced image can be calculated to determine the feasibility of each block size.

[0055] Specifically, the feasibility of each block size satisfies the following relationship: ; In the formula, is the feasibility of the th block size; is the local entropy of the th pixel in the enhanced image under the th block size; is the direction consistency of the th pixel in the enhanced image under the th block size; is the number of pixels in the enhanced image under the th block size; is the summation symbol.

[0056] Among them, reflects the noise confidence weight of the th pixel in the enhanced image under the th block size. When the direction consistency of all pixels with relatively high noise confidence weights in the enhanced image under the th block size is also relatively high, it indicates that image enhancement according to this block size is more appropriate, and the corresponding feasibility of this block size is relatively high.

[0057] Thus far, the determination of the feasibility of each block size can be completed.

[0058] S4: Convert the feasibility degree into weight coefficients with a sum of 1, and based on the weight coefficients, perform weighted summation on the gray values at the same pixel position in the enhanced images under all block sizes to obtain the target enhanced image, so as to detect the abnormal temperature on the surface of livestock and poultry based on the target enhanced image.

[0059] It should be noted that when performing image enhancement with a larger block size, the overall temperature distribution characteristics can be effectively retained; while when performing image enhancement with a smaller block size, the local details can be better highlighted. Therefore, in order to give full play to the advantages of both, the present invention performs dynamic weight allocation on the gray values at the same pixel position in the enhanced images under different block sizes according to the feasibility degree of the corresponding block size, and generates a comprehensive enhanced image, that is, the target enhanced image, through weighted fusion, so as to realize the accurate detection of the abnormal temperature on the surface of livestock and poultry based on the target enhanced image.

[0060] It should be further noted that in order to eliminate the influence of the absolute value difference of the feasibility degree, it is necessary to first convert the feasibility degree of each block size into weight coefficients.

[0061] Specifically, converting the feasibility degree of each block size into weight coefficients satisfies the following relational expression: ; In the formula, is the weight coefficient of the th block size; is the feasibility degree of the th block size; is the activation function, which is used to convert the feasibility degrees of all block sizes into weight coefficients with a sum of 1.

[0062] In another embodiment, the feasibility degrees of all block sizes can also be converted into weight coefficients with a sum of 1 through a normalization function, such as the function.

[0063] Furthermore, after converting the feasibility degree of each block size into weight coefficients, the weighted cumulative sum of the gray values at the same pixel position in the enhanced images under all block sizes can be calculated, so as to obtain the fused gray value at each pixel position and realize the determination of the target enhanced image.

[0064] Specifically, the fused gray value at any pixel position satisfies the following relational expression: ; In the formula, is the fused gray value at the th pixel position; is the weight coefficient of the th block size; is the gray value of the th pixel position in the enhanced image under the th block size; is the ordinal number of the maximum block size; is the summation symbol.

[0065] So far, the fused gray value of each pixel position can be obtained, thereby forming a target enhanced image, providing an accurate temperature distribution basis for subsequent abnormal detection of the body surface temperature of livestock and poultry.

[0066] In an exemplary embodiment of the present invention, the identification of abnormal body surface temperature of livestock and poultry can be achieved through the following steps: Obtain a preset gray threshold. If the gray value of any pixel point in the target enhanced image is less than the gray threshold, it is determined that the area where the pixel point is located is a normal temperature area; If it is greater than or equal to the gray threshold, the area where the pixel point is located is marked as an abnormal temperature area.

[0067] Optionally, the gray threshold can be set to 200 (gray value). If the gray value of a certain position in the target enhanced image is greater than the gray threshold, it is determined that the position exceeds the normal temperature range, and the position is marked. By analyzing and judging the gray values of all pixel points in the entire target enhanced image, the abnormal body temperature area on the body surface of livestock and poultry can be accurately identified, providing a reliable basis for livestock and poultry health monitoring and disease prevention and control.

[0068] It should be understood that various alternative solutions to the embodiments of the present invention described herein can be adopted in the process of practicing the present invention.

Claims

1. An abnormal body temperature detection method based on infrared thermal imaging, characterized in that, Including: Obtain the temperature distribution map of the livestock and poultry body surface and perform preprocessing to obtain an initial image; Based on different block sizes, use the CLAHE algorithm to perform image enhancement on the initial image respectively, calculate the direction consistency of each pixel point in the enhanced image under each block size, and the direction consistency represents the consistency between the gradient direction of the pixel point and the gradient directions of the corresponding pixel points in different enhanced images; Quantify the noise confidence weight of each pixel point in the enhanced image under each block size through the information entropy of all pixel points within the neighborhood range of the pixel point, and calculate the average value of the product of the noise confidence weights and the direction consistency of all pixel points in the same enhanced image to obtain the feasibility of each block size; Convert the feasibility into weight coefficients with a sum of 1, and according to the weight coefficients, perform weighted summation on the gray values of the same pixel position in the enhanced images under all block sizes to obtain a target enhanced image, so as to detect the abnormal temperature of the livestock and poultry body surface based on the target enhanced image.

2. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 1, wherein The calculating the direction consistency of each pixel point in the enhanced image under each block size includes: For the enhanced image under each block size, calculate the direction consistency of any pixel point according to the difference between the gradient direction angle of any pixel point in any enhanced image and the gradient direction angles of the corresponding pixel points in all the remaining enhanced images, and the direction consistency is negatively correlated with the difference.

3. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 2, wherein, The difference is a comprehensive difference, and the comprehensive difference satisfies the following relational expression: ； Wherein, is the comprehensive difference between the gradient direction angles of the th pixel in the enhanced image under the th block size and the gradient direction angle of the th pixel in the enhanced image under the th block size; , are respectively the size of the th block size and the size of the th block size; is the size of the maximum block size; is the ordinal number of the maximum block size; , are respectively the gradient direction angles of the th pixel in the enhanced image under the th block size and the th pixel in the enhanced image under the is the absolute value symbol; is a function that returns the minimum value; is the natural exponential function.

4. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 3, characterized in that, The calculating the direction consistency of any pixel point satisfies the following relational expression: ； In the formula, is the direction consistency of the -th pixel in the enhanced image under the -th block size; is the comprehensive difference between the gradient direction angle of the -th pixel and the gradient direction angle of the -th pixel in the enhanced image under the -th block size and the -th pixel in the enhanced image under the is the ordinal number of the maximum block size; is the natural exponential function.

5. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 4, wherein The method for obtaining the gradient direction angle includes: Use the Sobel operator to obtain the horizontal gradient value and the vertical gradient value of any pixel point, and calculate the arctangent value of the ratio of the vertical gradient value to the horizontal gradient value to obtain the gradient direction angle of any pixel point.

6. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 4, wherein The size of the maximum block size is the largest positive integer less than half of the width of the initial image, and the difference in the sizes of the block sizes with adjacent ordinals is a preset integer multiple of 2.

7. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 1, characterized in that The quantifying the noise confidence weight of each pixel point in the enhanced image under each block size through the information entropy of all pixel points within the neighborhood range of the pixel point includes: For each pixel point in the enhanced image under each block size, use the information entropy of all pixel points within the neighborhood range of any pixel point as the local entropy of any pixel point, and use the difference between 1 and the local entropy as the noise confidence weight of any pixel point.

8. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 7, wherein, The feasibility of each block size satisfies the following relational expression: ； Wherein, is the feasibility of the th block size; is the local entropy of the th pixel in the enhanced image under the th block size; is the directional consistency of the th pixel in the enhanced image under the th block size; is the number of pixels in the enhanced image under the th block size.

9. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 1, characterized in that The detecting the abnormal temperature of the livestock and poultry body surface based on the target enhanced image includes: Obtain a preset gray threshold. If the gray value of any pixel point in the target enhanced image is less than the gray threshold, it is determined that the area where the pixel point is located is a normal temperature area; If it is greater than or equal to the gray threshold, mark the area where the pixel point is located as an abnormal temperature area.

10. The method for detecting abnormal body temperature based on infrared thermal imaging according to claim 1, wherein, The method for obtaining the initial image includes: Perform grayscale processing and filtering processing on the temperature distribution map in sequence, and obtain the initial image after the filtering is completed.

Images