Infrared weak and small target detection method based on local contrast difference

Abstract

The invention belongs to the technical field of infrared image processing and target detection, and particularly relates to an infrared weak and small target detection method based on local contrast difference. Comprising the following steps: S1, obtaining an original infrared image, and carrying out mean filtering processing and background suppression preprocessing on the original infrared image to obtain a background suppression image; s2, designing a local contrast difference window, and processing the background suppression image by using the local contrast difference window to obtain a contrast difference image and a contrast difference enhancement coefficient matrix; and S3, calculating a confidence map based on the contrast difference image and the contrast difference enhancement coefficient matrix, and taking a pixel position corresponding to a gray scale maximum value in the confidence map as a detection position of the infrared weak and small target. According to the method, the pixel distribution coordinates of the infrared small target can be directly given, and the detection precision and the detection efficiency can be greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of infrared image processing and target detection, and particularly relates to an infrared small and weak target detection method based on local contrast difference. Background Art

[0002] The technology of detecting small and weak targets based on infrared images is an important part of infrared search and tracking systems and has a wide range of applications in fields such as early warning, missile guidance, and long-range surveillance. Therefore, it has always been one of the hot research fields competed by countries around the world. On the one hand, due to the relatively low resolution of infrared images, the size of small and weak targets in infrared images is small, and they lack rich texture and structure information. On the other hand, because the shooting scene is usually complex, small and weak targets are submerged in various backgrounds, resulting in a low signal-to-noise ratio. These complex factors make it difficult to accurately and effectively detect and extract infrared small and weak targets. Although many achievements have been made in the research on infrared small and weak target detection, further in-depth research is still needed to further enhance the ability to lock and track enemy targets and effectively enhance the awareness of the battlefield situation.

[0003] According to the form of the input image of the infrared small and weak target detection algorithm, existing algorithms can be divided into two categories, namely single-frame image detection algorithms and continuous multi-frame image detection algorithms. The input of the continuous multi-frame image detection algorithm is generally an infrared image sequence taken continuously with continuous target trajectories and small background change differences. This type of algorithm can combine the prior information of the input image to effectively achieve the trajectory estimation of the target. However, this type of algorithm has a large amount of calculation and a high calculation complexity, and has certain requirements for the computing platform resources of the system. The single-frame image detection algorithm is not restricted by the continuity of target and background changes. The structure of this type of algorithm is concise and clear, and has high timeliness. It can quickly and effectively estimate the target position according to a single infrared image. For systems with limited computing resources and high real-time requirements, the research on infrared small and weak target detection based on single-frame images has gradually become the mainstream. Currently, the research on the vast majority of infrared small and weak target detection algorithms is based on single-frame images. Based on the theoretical methods of the small and weak target detection algorithm for single-frame infrared images, it can be divided into detection methods based on filtering, detection algorithms based on statistical models, contrast enhancement algorithms based on human visual mechanisms, and deep learning algorithms based on big data, etc. The mechanisms of the first two detection algorithms are clear, relatively simple, and easy to implement in engineering. However, the detection probability decreases and the false alarm is high in complex backgrounds. The fourth detection algorithm has a high dependence on data, and the deep learning network is generally complex and not easy to implement in engineering. The third detection algorithm has good detection performance and real-time performance. However, when there are interferences such as high-brightness edges and high-brightness complex backgrounds in the scene, this type of algorithm generally cannot effectively suppress the interference. To obtain better detection results, this type of algorithm needs to perform calculations at multiple different scales. Summary of the Invention

[0004] In view of this, the present invention aims to provide an infrared dim and small target detection method based on local contrast difference to solve the problem that existing algorithms have poor ability to suppress complex background interference. The present invention can directly give the pixel distribution coordinates of infrared small targets, greatly improve the detection accuracy and detection efficiency, further improve the detection performance, algorithm real-time performance and reduce the false alarm probability.

[0005] To achieve the above object, the technical solution of the present invention is realized as follows: An infrared dim and small target detection method based on local contrast difference specifically includes the following steps: S1: Obtain the original infrared image, perform mean filtering processing and background suppression preprocessing on the original infrared image to obtain a background suppression image; S2: Design a local contrast difference window, and use the local contrast difference window to process the background suppression image to obtain a contrast difference image and a contrast difference enhancement coefficient matrix; S3: Calculate a confidence map based on the contrast difference image and the contrast difference enhancement coefficient matrix, and take the pixel position corresponding to the maximum gray value in the confidence map as the detection position of the infrared dim and small target.

[0006] Further, step S1 specifically includes the following steps: S11: Perform mean filtering processing on the original infrared image to obtain a filtered image; S12: Perform a difference processing on the original infrared image and the filtered image to obtain a difference image; S13: Use a background suppression operator to perform background suppression preprocessing on the difference image to obtain a background suppression image.

[0007] Further, in step S11, the scale of the mean filtering coefficient used in the mean filtering processing is 15×15.

[0008] Further, in step S13, the expression of the background suppression operator DF is: .

[0009] Further, in step S2, the local contrast difference window includes 5×5 sub-windows, the size of each sub-window is 3×3, the sub-window located at the center position of the local contrast difference window is used as the target sub-window, and the remaining sub-windows in the local contrast difference window except the target sub-window are used as background sub-windows; Number each sub-window in the order from top to bottom and from left to right. The expression of the average gray value of each sub-window is: ; Where, is the average gray value of the sub-window numbered i, N is the total number of pixels in the sub-window, and j is the j-th pixel in the sub-window. is the gray value of the j-th pixel in the sub-window numbered i in the background suppression image.

[0010] Further, step S2 specifically includes the following steps: S21: Based on ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), calculate the minimum contrast difference of the target sub-window to obtain the first contrast difference image : ; Among them, is the contrast difference calculated by the target sub-window based on ( , , , ) and ( , , , ), is the contrast difference calculated by the target sub-window based on ( , , , ) and ( , , , ), The contrast difference calculated for the target sub-window based on ( , , , ) and ( , , , ); The contrast difference calculated for the target sub-window based on ( , , , ) and ( , , , ); S22: Calculate the contrast differences of the target sub-window in the horizontal direction, vertical direction, 45° direction, and 135° direction, and use the minimum value of the contrast differences of the target sub-window in the four directions as the second contrast difference image : ; Wherein, is the contrast difference of the target sub-window in the horizontal direction, is the contrast difference of the target sub-window in the vertical direction, is the contrast difference of the target sub-window in the 45° direction, is the contrast difference of the target sub-window in the 135° direction; S23: Based on the first contrast difference image and the second contrast difference image , calculate the contrast difference image DG : ; S24: Divide the background sub-window into 8 regions, and the 8 regions are respectively: ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), and calculate the maximum mean gray value of each region: ; Among them, is the h-th region, h ≤ 8, , , and respectively represent the average gray values of the four sub-windows within the region, is to take the maximum value among the average gray values of the four sub-windows as the maximum mean gray value of the region; S25: Sort the maximum mean gray values of the 8 regions in descending order, take the average of the maximum mean gray values ranked in the top four as the target average value, and calculate the difference between the square of the average gray value of the target sub-window and the square of the target average value to obtain the contrast difference enhancement coefficient matrix.

[0011] Furthermore, step S21 includes the following steps: S211: Calculate the average gray values of the sub-windows on the left and right sides of the target sub-window based on ( , , , ) and ( , , , ) to obtain the contrast difference : ; ; ; Among them, is the average gray value of the four sub-windows of ( , , , ), is the average gray value of the four sub-windows of ( , , , ); S212: Based on ( , , , ), and ( , , , ) calculate the grayscale average values of the upper and lower sub - windows of the target sub - window to obtain the contrast difference : ; ; ; Among them, is the grayscale average value of these four sub - windows ( , , , ), is the grayscale average value of these four sub - windows ( , , , ); S213: Based on ( , , , ) and ( , , , ) calculate the grayscale average values of the upper and lower sub - windows of the target sub - window to obtain the contrast difference : ; ; ; Among them, is the grayscale average value of these four sub - windows ( , , , ), is the grayscale average value of these four sub - windows ( , , , ); S214: Based on ( , , , ) and ( , , , Calculate the grayscale average values of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ; ; ; Among them, is the grayscale average value of these four sub-windows of ( , , , ), is the grayscale average value of these four sub-windows of ( , , , ).

[0012] Furthermore, step S22 includes the following steps: S221: Calculate the contrast difference of the target sub-window in the horizontal direction : ; ; ; Among them, is the grayscale average value of one side of the target sub-window in the horizontal direction, is the grayscale average value of the other side of the target sub-window in the horizontal direction; S222: Calculate the contrast difference of the target sub-window in the vertical direction : ; ; ; Among them, is the grayscale average value of one side of the target sub-window in the vertical direction, is the grayscale average value of the other side of the target sub-window in the vertical direction; S223: Calculate the contrast difference of the target sub-window in the 45° direction : ; ; ; Among them, is the grayscale average value of one side of the target sub-window in the 45° direction, is the grayscale average value of the other side of the target sub-window in the 45° direction; S224: Calculate the contrast difference of the target sub-window in the 135° direction : ; ; ; Among them, is the average gray value on one side of the target sub-window in the 135° direction, is the average gray value on the other side of the target sub-window in the 135° direction.

[0013] Furthermore, in step S3, the calculation formula of the confidence map R is: ; Among them, is the contrast difference enhancement coefficient matrix, DG is the contrast difference image.

[0014] Compared with the prior art, the present invention can achieve the following beneficial effects: (1) The infrared small and weak target detection method based on local contrast difference according to the present invention can directly give the pixel distribution coordinates of the infrared small target, and can greatly improve the detection accuracy and detection efficiency. While effectively suppressing background interference, it can further improve the detection performance, algorithm real-time performance and reduce the false alarm probability.

[0015] (2) For the infrared small and weak target detection method based on local contrast difference according to the present invention, the designed local contrast difference window of the present invention includes both the background around the target and the outermost far background. Using the local contrast difference window to process the background suppression image, its background suppression effect will be greatly enhanced, improving the accuracy of infrared small and weak target detection and reducing the calculation cost.

[0016] (3) The infrared small and weak target detection method based on local contrast difference according to the present invention uses the contrast difference enhancement coefficient matrix to further enhance the target contrast and effectively reduce background interference. Description of the Drawings

[0017] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings: Figure 1 is the flowchart of the infrared small and weak target detection method based on local contrast difference according to the embodiment of the present invention; Figure 2The original infrared image and the 3D diagram of the original infrared image described in the embodiment of the present invention; Figure 3 The difference image and the 3D diagram of the difference image described in the embodiment of the present invention; Figure 4 The background suppression image and the 3D diagram of the background suppression image described in the embodiment of the present invention; Figure 5 The structural schematic diagram of the local contrast difference window described in the embodiment of the present invention; Figure 6 The first contrast difference image and the 3D diagram of the first contrast difference image described in the embodiment of the present invention; Figure 7 The second contrast difference image and the 3D diagram of the second contrast difference image described in the embodiment of the present invention; Figure 8 The contrast difference image and the 3D diagram of the contrast difference image described in the embodiment of the present invention; Figure 9 The 2D diagram and the 3D diagram of the contrast difference enhancement coefficient matrix described in the embodiment of the present invention; Figure 10 The confidence map and the 3D diagram of the confidence map described in the embodiment of the present invention. Detailed implementation manners

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not constitute a limitation to the present invention.

[0019] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0020] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first", "second", etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise stated, the meaning of "a plurality of" is two or more.

[0021] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "mounted", "connected", "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood through specific situations.

[0022] The present invention will be described in detail below with reference to the drawings and in conjunction with embodiments.

[0023] As Figure 1 shown, the present invention provides an infrared dim small target detection method based on local contrast difference, which specifically includes the following steps: S1: Obtain the original infrared image, perform mean filtering processing and background suppression preprocessing on the original infrared image to obtain a background suppression image; S2: Design a local contrast difference window, and use the local contrast difference window to process the background suppression image to obtain a contrast difference image and a contrast difference enhancement coefficient matrix; S3: Calculate a confidence map based on the contrast difference image and the contrast difference enhancement coefficient matrix, and use the pixel position corresponding to the maximum gray value in the confidence map as the detection position of the infrared dim small target.

[0024] It should be noted that first, calculate two types of local contrast differences based on the double-layer background region, and fuse the results of the two types of local contrast differences; then, obtain the contrast difference enhancement coefficient matrix according to the mean brightness difference of the double-layer background region; finally, obtain the estimation result of the infrared dim small target according to the fusion result and the local contrast coefficient.

[0025] In some embodiments, step S1 specifically includes the following steps: S11: Perform mean filtering on the original infrared image to obtain a filtered image; S12: Perform a difference operation on the original infrared image and the filtered image to obtain a difference image; S13: Use a background suppression operator to perform background suppression preprocessing on the difference image to obtain a background suppression image.

[0026] In some embodiments, in step S11, the scale of the mean filtering coefficient used for mean filtering is 15×15.

[0027] It should be noted that the scale of the mean filtering coefficient is the same as the scale of the local contrast difference window.

[0028] Further, as Figures 2 - 4 shown, in step S12, the original infrared image and the filtered image are subjected to a difference operation to obtain a difference image : ; Use a symmetric background suppression operator DF with a scale of to perform background suppression preprocessing on the difference image to obtain a preprocessed background suppression image , and the calculation process is shown in the following formula: ; where p and q are the horizontal and vertical coordinates of the DF operator.

[0029] In some embodiments, in step S13, the expression of the background suppression operator DF is: .

[0030] The local contrast difference window includes 5×5 sub-windows, each sub-window has a size of 3×3. The sub-window located at the center position of the local contrast difference window is used as the target sub-window, and the remaining sub-windows in the local contrast difference window except the target sub-window are used as background sub-windows; Number the sub-windows in the order from top to bottom and from left to right. The expression for the average gray value of each sub-window is: ; where is the average gray value of the sub-window numbered i, i = 1, 2, 3,..., 25; N is the total number of pixels in the sub-window, N = 1, 2, 3,..., 9; j is the jth pixel in the sub-window, is the gray value of the jth pixel in the sub-window numbered i in the background suppression image.

[0031] It should be noted that, as Figure 5 shown, since the local contrast difference window includes 5×5 sub-windows, and the size of each sub-window is 3×3, therefore, the scale of the local contrast difference window is ; where the central window C represents the target sub-window, and the peripheral window B represents the background sub-window.

[0032] In some embodiments, as Figures 6 - 9 shown, step S2 specifically includes the following steps: S21: Based on ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), calculate the minimum value of the contrast difference of the target sub-window to obtain the first contrast difference image : ; Wherein, is the contrast difference calculated by the target sub-window based on ( , , , ) and ( , , , ), is the contrast difference calculated by the target sub-window based on ( , , , ) and ( , , , The calculated contrast difference, is the contrast difference calculated for the target sub-window based on ( , , , ) and ( , , , ); is the contrast difference calculated for the target sub-window based on ( , , , ) and ( , , , ); S22: Calculate the contrast differences of the target sub-window in the horizontal, vertical, 45°, and 135° directions, and take the minimum value of the contrast differences of the target sub-window in the four directions as the second contrast difference image : ; wherein, is the contrast difference of the target sub-window in the horizontal direction, is the contrast difference of the target sub-window in the vertical direction, is the contrast difference of the target sub-window in the 45° direction, is the contrast difference of the target sub-window in the 135° direction; S23: Based on the first contrast difference image and the second contrast difference image , calculate the contrast difference image DG : ; S24: Divide the background sub-window into 8 regions, and the 8 regions are respectively: ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), ( , , , ), and calculate the maximum mean gray value of each region: ; Among them, is the h-th region, h ≤ 8, , , and respectively represent the average gray values of the four sub-windows within the region, is to take the maximum value among the average gray values of the four sub-windows as the maximum mean gray value of the region; S25: Sort the maximum mean gray values of the 8 regions in descending order, take the average of the top four maximum mean gray values as the target average value, and subtract the square of the average gray value of the target sub-window from the square of the target average value to obtain the contrast difference enhancement coefficient matrix.

[0033] It should be noted that the main function of the contrast difference enhancement coefficient matrix is: to further enhance the weak targets in DG the contrast difference image and suppress the influence of non-target regions. The specific calculation process is as follows: For different sub-window regions around the target region C, calculate the maximum mean gray value within this region. The calculation formula is as follows: ; ; ; ; ; ; ; ; Then, sort the above 8 maximum mean gray values, and take the top 4 values from largest to smallest, which are respectively denoted as , , and , the contrast difference enhancement coefficient matrix PG can be expressed as: .

[0034] Further, in step S23, the contrast difference image DG is expressed as the fusion of the contrast difference image and the contrast difference image .

[0035] In some embodiments, step S21 includes the following steps: S211: Based on ( , , , ) and ( , , , ) calculate the grayscale average values of the sub-windows on the left and right sides of the target sub-window to obtain the contrast difference : ; ; ; wherein, is the grayscale average value of these four sub-windows of ( , , , ), is the grayscale average value of these four sub-windows of ( , , , ); S212: Based on ( , , , ) and ( , , , ) calculate the grayscale average values of the sub-windows on the upper and lower sides of the target sub-window to obtain the contrast difference : ; ; ; wherein, is the grayscale average value of these four sub-windows of ( , , , ), is ( , , , ) the grayscale average values of these four sub-windows; S213: Based on ( , , , ) and ( , , , ) calculate the grayscale average values of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ; ; ; Wherein, is the grayscale average value of these four sub-windows of ( , , , ), is the grayscale average value of these four sub-windows of ( , , , ); S214: Based on ( , , , ) and ( , , , ) calculate the grayscale average values of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ; ; ; Wherein, is the grayscale average value of these four sub-windows of ( , , , ), is the grayscale average value of these four sub-windows of ( , , , ).

[0036] Based on the contrast difference and the contrast difference , contrast difference and contrast difference , and calculate to obtain a first contrast difference image : .

[0037] In one embodiment, step S22 includes the following steps: S221: Calculate the contrast difference of the target sub-window in the horizontal direction : ; ; ; wherein, is the average gray value of one side of the target sub-window in the horizontal direction, is the average gray value of the other side of the target sub-window in the horizontal direction; S222: Calculate the contrast difference of the target sub-window in the vertical direction : ; ; ; wherein, is the average gray value of one side of the target sub-window in the vertical direction, is the average gray value of the other side of the target sub-window in the vertical direction; S223: Calculate the contrast difference of the target sub-window in the 45° direction : ; ; ; wherein, is the average gray value of one side of the target sub-window in the 45° direction, is the average gray value of the other side of the target sub-window in the 45° direction; S224: Calculate the contrast difference of the target sub-window in the 135° direction : ; ; ; wherein, is the average gray value of one side of the target sub-window in the 135° direction, is the average gray value on the other side of the target sub-window in the 135° direction.

[0038] Based on the contrast difference , the contrast difference , the contrast difference and the contrast difference , calculate to obtain the second contrast difference image : .

[0039] It should be noted that, different from calculating , obtaining only considers the sub-windows in a certain direction and does not consider the sub-windows on both sides of the direction axis. The average gray value of the pixels of the sub-window where the target region C is located is , and use to calculate the contrast difference in each direction.

[0040] In some embodiments, in step S3, the calculation formula of the confidence map R is: ; where is the contrast difference enhancement coefficient matrix, DG is the contrast difference image.

[0041] It should be noted that, as Figure 10 shown, according to the maximum value position in the confidence map R, the detection position of the infrared small and weak target can be obtained.

[0042] It should be understood that various forms of the flow shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure of the invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and no limitation is made herein.

[0043] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for detecting infrared small targets based on local contrast differences, characterized in that: The specific steps include: S1: Acquire an original infrared image, perform mean filtering and background suppression preprocessing on the original infrared image, and obtain a background suppression image; S2: Design a local contrast difference window, and use the local contrast difference window to process the background suppression image to obtain a contrast difference image and a contrast difference enhancement coefficient matrix; S3: Calculate the confidence map based on the contrast difference image and the contrast difference enhancement coefficient matrix, and use the pixel position corresponding to the grayscale maximum value in the confidence map as the detection position of the infrared weak target.

2. The infrared small target detection method based on local contrast difference according to claim 1 is characterized in that: Step S1 specifically includes the following steps: S11: Perform mean filtering on the original infrared image to obtain a filtered image; S12: performing difference processing on the original infrared image and the filtered image to obtain a difference image; S13: Perform background suppression preprocessing on the difference image using a background suppression operator to obtain a background suppressed image.

3. The infrared small target detection method based on local contrast difference according to claim 2 is characterized in that: In step S11, the scale of the mean filter coefficients used in the mean filter process is 15×15.

4. The infrared small target detection method based on local contrast difference according to claim 2 is characterized in that: In step S13, the expression of the background suppression operator DF is: 。 5. The infrared small target detection method based on local contrast difference according to claim 1 is characterized in that: In step S2, the local contrast difference window includes 5×5 sub-windows, each sub-window has a size of 3×3, the sub-window located at the center of the local contrast difference window is used as the target sub-window, and the remaining sub-windows in the local contrast difference window except the target sub-window are used as background sub-windows; The sub-windows are numbered from top to bottom and from left to right. The expression of the average gray value of each sub-window is: ； in, is the average gray value of the subwindow numbered i, N is the total number of pixels in the subwindow, j is the jth pixel in the subwindow, is the gray value of the jth pixel in the subwindow numbered i in the background suppressed image.

6. The infrared small target detection method based on local contrast difference according to claim 5 is characterized in that: Step S2 specifically includes the following steps: S21: Based on ( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , ), calculate the minimum contrast difference of the target sub-window, and obtain the first contrast difference image : ； in, The target subwindow is based on ( , , , )and( , , , ) is the calculated contrast difference, The target subwindow is based on ( , , , )and( , , , ) is the calculated contrast difference, The target subwindow is based on ( , , , )and( , , , ) is the calculated contrast difference, The target subwindow is based on ( , , , )and( , , , ) The calculated contrast difference; S22: Calculate the contrast difference of the target sub-window in the horizontal direction, vertical direction, 45° direction and 135° direction, and take the minimum contrast difference of the target sub-window in the four directions as the second contrast difference image : ； in, is the contrast difference of the target sub-window in the horizontal direction, is the contrast difference of the target sub-window in the vertical direction, is the contrast difference of the target sub-window at 45°, is the contrast difference of the target sub-window in the 135° direction; S23: Based on the first contrast difference image and the second contrast difference image , the contrast difference image is calculated DG : ； S24: Divide the background sub-window into 8 regions, the 8 regions are: ( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , )、( , , , ), and calculate the mean grayscale maximum value of each region: ； in, is the hth region, h≤8, , , and Respectively represent the average grayscale values ​​of the four background sub-windows in the area, The maximum value among the average grayscale values ​​of the four background sub-windows is taken as the average grayscale maximum value of the region; S25: Sort the mean grayscale maximum values ​​of the eight regions in descending order, take the average of the mean grayscale maximum values ​​in the first four positions as the target average value, and make a difference between the square of the mean grayscale value of the target sub-window and the square of the target average value to obtain a contrast difference enhancement coefficient matrix.

7. The infrared small target detection method based on local contrast difference according to claim 6 is characterized in that: Step S21 includes the following steps: S211: Based on ( , , , )and( , , , ) Calculate the grayscale average of the left and right sub-windows of the target sub-window to obtain the contrast difference : ； ； ； in, for( , , , ) The grayscale average of these four sub-windows, for( , , , ) The grayscale average of these four sub-windows; S212: Based on ( , , , )and( , , , ) Calculate the grayscale average of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ； ； ； in, for( , , , ) The grayscale average of these four sub-windows, for( , , , ) The grayscale average of these four sub-windows; S213: Based on ( , , , )and( , , , ) Calculate the grayscale average of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ； ； ； in, for( , , , ) The grayscale average of these four sub-windows, for( , , , ) The grayscale average of these four sub-windows; S214: Based on ( , , , )and( , , , ) Calculate the grayscale average of the upper and lower sub-windows of the target sub-window to obtain the contrast difference : ； ； ； in, for( , , , ) The grayscale average of these four sub-windows, for( , , , ) is the grayscale average of these four sub-windows.

8. The infrared small target detection method based on local contrast difference according to claim 6 is characterized in that: Step S22 includes the following steps: S221: Calculate the contrast difference of the target sub-window in the horizontal direction : ； ； ； in, is the grayscale average of the target sub-window on one side in the horizontal direction, is the grayscale average value of the target sub-window on the other side in the horizontal direction; S222: Calculate the contrast difference of the target sub-window in the vertical direction : ； ； ； in, is the grayscale average of the target sub-window on one side in the vertical direction, is the grayscale average value of the target sub-window on the other side in the vertical direction; S223: Calculate the contrast difference of the target sub-window in the 45° direction : ； ； ； in, is the grayscale average of the target sub-window on one side of the 45° direction, is the grayscale average of the target sub-window on the other side of the 45° direction; S224: Calculate the contrast difference of the target sub-window in the 135° direction : ； ； ； in, is the grayscale average of the target sub-window on one side of the 135° direction, It is the grayscale average of the target sub-window on the other side of the 135° direction.

9. The infrared small target detection method based on local contrast difference according to claim 1 is characterized in that: In step S3, the confidence map R is calculated as: ； in, is the contrast difference enhancement coefficient matrix, DG is the contrast difference image.

Images