Infrared dim small target detection method based on local mean difference minimum value

Abstract

The present application belongs to the technical field of infrared image target detection, and particularly relates to an infrared weak small target detection method based on local mean difference minimum value. The method comprises the following steps: S1: performing mean filtering processing on the original infrared image to obtain a background suppression image; S2: numbering each sub-window in the order of column snake arrangement, sequentially setting a local mean difference acquisition window, and obtaining a mean filtering result with the target sub-window as the center corresponding to the setting result; S3: based on the mean filtering result of each step S2, calculating a first mean difference image and a second mean difference image; S4: performing point multiplication operation on the first mean difference image and the second mean difference image to obtain a local mean difference minimum value confidence image, and taking the maximum value point position in the local mean difference minimum value confidence image as the position of the infrared weak small target. The present application can adapt to different target scales and solve the problem of large amount of calculation caused by multi-scale detection.

Description

Technical Field

[0001] This invention belongs to the technical field of infrared image target detection, and particularly relates to a method for detecting weak infrared targets based on the minimum difference of local mean values. Background Technology

[0002] Infrared imaging detection is only related to the temperature and material properties of the object being detected, and has unique advantages compared with visible light imaging and radar imaging. Infrared imaging detection can work in low light conditions and is not affected by light intensity. At the same time, infrared imaging detection has strong penetration ability, is a passive detection method, is not easy to expose its own position, and is not easily interfered with by other parties. During the infrared detection process, as the detection distance increases, the target in the captured image is very small in pixel size and does not have shape and texture features, existing as a weak target. The weak target detection technology of infrared image is one of the core technologies of infrared search and track systems, and its detection performance directly affects the system's ability to track targets, predict trajectories, and acquire motion orientation information.

[0003] When acquiring infrared targets, the long detection distance makes the target appear very small in the infrared image, lacking shape and texture features. Furthermore, long-distance detection introduces various noises that can overlap, causing the target to blend into the background and resulting in a low signal-to-noise ratio. In addition, the imaging scene is generally complex, making the background and target highly similar in the image, which can easily lead to false detections by the detection algorithm. These complex factors make it difficult to detect small infrared targets with high accuracy.

[0004] Compared to detection algorithms based on infrared image sequences, weak target detection algorithms based on single-frame infrared images are simpler, have lower computational cost, and offer higher real-time performance, requiring less computational platform resources. Therefore, they can capture target information in real time and quickly assist the system in decision-making. Local contrast difference algorithms based on human visual system models are an emerging class of infrared weak target detection algorithms. These algorithms can enhance the local contrast features of the target, thereby achieving target detection. However, in practical applications, these algorithms generally have poor adaptability when the target scale changes, easily leading to false detections and alarms. Although some researchers have proposed multi-scale detection algorithms, these require complex extremum searches in multi-scale spaces, resulting in high computational cost and demanding certain computational platform resources. Summary of the Invention

[0005] In view of this, the present invention aims to provide an infrared weak target detection method based on the minimum value of local mean difference, so as to solve the problem of large computational load when the target scale is transformed by existing local contrast difference algorithms, and further improve the detection efficiency of the algorithm, reduce false alarms, and improve the real-time performance of the algorithm. The present invention can adapt to different target scales and solve the problem of large computational load caused by multi-scale detection.

[0006] To achieve the above objectives, the technical solution created by this invention is implemented as follows:

[0007] A method for detecting weak infrared targets based on minimizing local mean differences, specifically including the following steps:

[0008] S1: Perform mean filtering on the original infrared image to obtain a background-suppressed image;

[0009] S2: Construct a local mean difference acquisition window, take the central sub-window of the local mean difference acquisition window as the target sub-window, number each sub-window in a column serpentine arrangement, and set the local mean difference acquisition window in descending order of sub-window number according to the filter setting rules to obtain the mean filtering result centered on the target sub-window corresponding to the setting result;

[0010] The filter setting rule is: set the current sub-window to the mean filter, and set all elements of the other sub-windows to 0;

[0011] S3: Based on the mean filtering results of step S2, calculate the first mean difference image and the second mean difference image accordingly;

[0012] S4: Perform a dot product operation on the first mean difference image and the second mean difference image to obtain a local mean difference minimum confidence image, and take the location of the maximum point in the local mean difference minimum confidence image as the location of the infrared weak target.

[0013] Furthermore, step S1 specifically includes:

[0014] S11: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the first filtered image. ;

[0015] S12: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the second filtered image. ;

[0016] S13: Subtract the first filtered image from the second filtered image to obtain the background-suppressed image. :

[0017] ;

[0018] in, These are the coordinates of the pixel's position on the image.

[0019] Furthermore, step S2 specifically includes the following steps:

[0020] S21: Local mean difference acquisition window includes Each sub-window measures 3×3.

[0021] S22: Set the sub-window numbered 1 as the mean filter, and set all elements of the remaining sub-windows to 0;

[0022] S23: Based on the current local mean difference, the background suppression image is traversed through the window, and the mean filtering result centered on the target sub-window is obtained by the following formula. :

[0023] ;

[0024] in, Background suppression image, These are the coordinates of the pixel's position on the image. To obtain a window for the current local mean difference, Obtain the location index of the window for the local mean difference;

[0025] S24: Repeat steps S22-S23 in ascending order of number to obtain the mean filtering result centered on the target sub-window, corresponding to the setting result.

[0026] Furthermore, in step S3, the specific steps for obtaining the first mean difference image are as follows:

[0027] S3A1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the first mean difference :

[0028] ;

[0029] in, The mean filtering result centered on the target sub-window, obtained under the condition that the target sub-window is set as the mean filter and all elements of the other sub-windows are set to 0; The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 7 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 19 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 25 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0030] S3A2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference between the second means :

[0031] ;

[0032] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 11 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 12 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 14 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 15 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0033] S3A3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the third mean :

[0034] ;

[0035] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 21 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 17 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 5 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 9 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0036] S3A4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fourth mean difference :

[0037] ;

[0038] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 18 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 23 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 3 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 8 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0039] S3A5: Based on the calculation results of steps S3A1-S3A4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. :

[0040]

[0041] in, These are the coordinates of the pixel's position on the image.

[0042] Furthermore, in step S3, the specific steps for obtaining the second mean difference image are as follows:

[0043] S3B1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fifth mean difference :

[0044] ;

[0045] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 6 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 16 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 10 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 20 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0046] S3B2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the sixth mean difference :

[0047] ;

[0048] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 22 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 4 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0049] S3B3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the seventh mean difference :

[0050] ;

[0051] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 24 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 2 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0052] S3B4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the eighth mean. :

[0053] ;

[0054] S3B5: Based on the calculation results of steps S3B1-S3B4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. :

[0055]

[0056] in, These are the coordinates of the pixel's position on the image.

[0057] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0058] The infrared weak target detection method based on the minimum difference of local mean, as described in this invention, can directly give the position of infrared small targets in infrared images, which can greatly improve detection accuracy and real-time performance. Attached Figure Description

[0059] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0060] Figure 1 A schematic flowchart of the infrared weak target detection method based on the minimum difference of local mean values ​​described in the embodiments of the present invention;

[0061] Figure 2(a) shows the original infrared input image according to the embodiment of the present invention;

[0062] Figure 2(b) shows the first filtered image according to an embodiment of the present invention;

[0063] Figure 2(c) shows the second filtered image according to an embodiment of the present invention;

[0064] Figure 2(d) is a background suppression image according to an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of the structure of the local mean difference acquisition window described in the embodiment of the present invention;

[0066] Figure 4 A schematic diagram illustrating the principle of obtaining the first mean difference as described in an embodiment of the present invention;

[0067] Figure 5 A schematic diagram illustrating the principle of obtaining the second mean difference as described in an embodiment of the present invention;

[0068] Figure 6 A three-dimensional grayscale image of the first mean difference image described in the embodiments of the present invention;

[0069] Figure 7 A schematic diagram illustrating the principle of obtaining the fifth mean difference as described in an embodiment of the present invention;

[0070] Figure 8 A schematic diagram illustrating the principle of obtaining the sixth mean difference as described in an embodiment of the present invention;

[0071] Figure 9 This invention creates a three-dimensional grayscale image of the second mean difference image as described in the embodiments of the present invention.

[0072] Figure 10This is a grayscale three-dimensional image of the confidence map of the minimum local mean difference described in the embodiments of the present invention. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of this invention clearer, the 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 merely illustrative of the invention and do not constitute a limitation thereof.

[0074] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0075] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0076] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0077] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0078] like Figure 1As shown, an infrared weak target detection method based on the minimum local mean difference includes the following steps: S1: Perform mean filtering on the original infrared image to obtain a background-suppressed image; S2: Construct a local mean difference acquisition window, take the central sub-window of the local mean difference acquisition window as the target sub-window, number each sub-window in a column serpentine arrangement, and set the local mean difference acquisition window in descending order of sub-window number according to the filter setting rule to obtain the mean filtering result centered on the target sub-window corresponding to the setting result; the filter setting rule is: set the current sub-window as the mean filter, and set all elements of the other sub-windows to 0; S3: Calculate the first mean difference image and the second mean difference image based on the mean filtering results of step S2; S4: Perform a dot product operation on the first mean difference image and the second mean difference image to obtain a local mean difference minimum confidence image, and take the maximum point in the local mean difference minimum confidence image as the location of the infrared weak target.

[0079] It should be noted that, firstly, the difference between the two local means of the image after background suppression processing is calculated; then, the two local mean difference images are fused; finally, the position estimate of the infrared small target is obtained based on the fusion result. This invention can directly give the position of the infrared small target in the infrared image, which can greatly improve the detection accuracy and real-time performance.

[0080] Furthermore, step S1 specifically includes:

[0081] S11: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the first filtered image. ;

[0082] S12: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the second filtered image. ;

[0083] S13: Subtract the first filtered image from the second filtered image to obtain the background-suppressed image. :

[0084] ;

[0085] in, These are the coordinates of the pixel's position on the image.

[0086] It should be noted that, as shown in Figures 2(a)-2(d), the original infrared input image Based on size Mean filter coefficients, applied to them for mean filtering, yield the filtered image. Based on size Mean filter coefficients, applied to them for mean filtering, yield the filtered image. The image after background suppression can be represented as:

[0087] ;

[0088] Furthermore, step S2 specifically includes the following steps:

[0089] S21: Local mean difference acquisition window includes Each sub-window measures 3×3.

[0090] S22: Set the sub-window numbered 1 as the mean filter, and set all elements of the remaining sub-windows to 0;

[0091] S23: Based on the current local mean difference, the background suppression image is traversed through the window, and the mean filtering result centered on the target sub-window is obtained by the following formula. :

[0092] ;

[0093] in, Background suppression image, These are the coordinates of the pixel's position on the image. To obtain a window for the current local mean difference, Obtain the location index of the window for the local mean difference;

[0094] S24: Repeat steps S22-S23 in ascending order of number to obtain the mean filtering result centered on the target sub-window, corresponding to the setting result.

[0095] It should be noted that, as Figure 3 As shown in the figure below, the local mean difference acquisition window of the infrared image target extraction algorithm contains 25 sub-windows. Arranged in a manner where each sub-window is [size missing]. The center sub-window is T, which represents the target sub-window.

[0096] Of the 25 sub-windows, only sub-window 1 uses... The mean filter of the size, when all elements in other sub-windows are 0, the above The local mean difference acquisition window is represented as Then, the image is centered on the sub-window T. Mean filtering results It can be represented as:

[0097] ;

[0098] Similarly, when only child window 2 is The mean filter of the size, when all elements in other sub-windows are 0, the above The local mean difference acquisition window is represented as Then, the image is centered on the sub-window T. Mean filtering results It can be represented as:

[0099] ;

[0100] By analogy, we can obtain... ~ , , ~ .

[0101] Furthermore, in step S3, the specific steps for obtaining the first mean difference image are as follows:

[0102] S3A1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the first mean difference :

[0103] ;

[0104] in, The mean filtering result centered on the target sub-window, obtained under the condition that the target sub-window is set as the mean filter and all elements of the other sub-windows are set to 0; The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 7 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 19 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 25 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0105] S3A2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference between the second means :

[0106] ;

[0107] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 11 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 12 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 14 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 15 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0108] S3A3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the third mean :

[0109] ;

[0110] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 21 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 17 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 5 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 9 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0111] S3A4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fourth mean difference :

[0112] ;

[0113] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 18 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 23 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 3 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 8 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0114] S3A5: Based on the calculation results of steps S3A1-S3A4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. :

[0115] .

[0116] It should be noted that, as Figure 4 As shown in the figure below, the local mean difference is obtained centered on sub-window T. The first mean difference... It can be represented as:

[0117] ;

[0118] like Figure 5 As shown, the second mean difference It can be represented as:

[0119] ;

[0120] Similarly, the third mean difference can be obtained by analogy. Difference from the fourth mean It can be expressed as the following formula:

[0121] ;

[0122] ;

[0123] For the above four mean difference images , , , At the same pixel position, take the minimum value among four pixels, such as Figure 6 As shown, the first mean difference image is obtained. , can be represented as:

[0124]

[0125] Furthermore, in step S3, the specific steps for obtaining the second mean difference image are as follows:

[0126] S3B1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fifth mean difference :

[0127] ;

[0128] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 6 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 16 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 10 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 20 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0129] S3B2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the sixth mean difference :

[0130] ;

[0131] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 22 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 4 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0132] S3B3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the seventh mean difference :

[0133] ;

[0134] in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 24 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 2 is set as the mean filter and all elements of the other sub-windows are set to 0.

[0135] S3B4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the eighth mean. :

[0136] ;

[0137] S3B5: Based on the calculation results of steps S3B1-S3B4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. :

[0138] .

[0139] It should be noted that, as Figure 7 As shown in the figure below, the local mean difference is obtained centered on sub-window T, as shown in the fifth mean difference. It can be represented as:

[0140] ;

[0141] like Figure 8 As shown, the sixth mean difference It can be represented as:

[0142] ;

[0143] Similarly, by analogy, we can obtain the seventh mean difference. Difference from the eighth mean It can be expressed as the following formula:

[0144] ;

[0145] ;

[0146] For the above four mean difference images , , , At the same pixel position, take the minimum value among four pixels, such as Figure 9 As shown, the second mean difference image is obtained. , can be represented as:

[0147] .

[0148] Finally, as Figure 10 As shown, the confidence image of the minimum local mean difference. It can be represented as:

[0149] ;

[0150] Among them, operators This represents the dot product operation where each pixel is multiplied; the confidence image is found at the local mean minimum difference. In the diagram, the location of the maximum value point indicates the location of the weak infrared target.

[0151] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0152] The specific embodiments described above do not constitute a limitation on the scope of protection of this 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 this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting an infrared dim small target based on a local mean difference minimum value, characterized in that: Specifically, the steps include the following: S1: Perform mean filtering on the original infrared image to obtain a background-suppressed image; S2: Construct a local mean difference acquisition window, take the central sub-window of the local mean difference acquisition window as the target sub-window, number each sub-window in a column serpentine arrangement, and set the local mean difference acquisition window in descending order of sub-window number according to the filter setting rules to obtain the mean filtering result centered on the target sub-window corresponding to the setting result; The filter setting rule is: set the current sub-window to the mean filter, and set all elements of the other sub-windows to 0; Step S2 specifically includes the following steps: S21: the local mean difference acquisition window contains sub-windows, and each sub-window has a size of 3x3; S22: Set the sub-window numbered 1 as the mean filter, and set all elements of the remaining sub-windows to 0; S23: Based on the current local mean difference acquisition window, the background suppression image is traversed, and the mean filtering result centered on the target sub-window is obtained by the following formula : ； wherein, is a background inhibited image, is a position coordinate of the pixel point on the image, is a current local mean difference acquisition window, is a position index of the local mean difference acquisition window; S24: Repeat steps S22-S23 in ascending order of number to obtain the mean filtering result centered on the target sub-window, which corresponds to the setting result; S3: Based on the mean filtering results of step S2, calculate the first mean difference image and the second mean difference image accordingly; In step S3, the specific steps for obtaining the first mean difference image are as follows: S3A1: based on the mean filtering result , the mean filtering result , the mean filtering result , the mean filtering result and the mean filtering result , calculate the first mean difference ; S3A2: based on the mean filtering result , the mean filtering result , the mean filtering result , the mean filtering result and the mean filtering result , calculate a second mean difference ; S3A3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the third mean ; S3A4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fourth mean difference ; S3A5: Based on the calculation results of steps S3A1-S3A4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. ; The mean filtering result centered on the target sub-window, obtained under the condition that the target sub-window is set as the mean filter and all elements of the other sub-windows are set to 0; The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered i is set as the mean filter and all elements of the other sub-windows are set to 0. In step S3, the specific steps for obtaining the second mean difference image are as follows: S3B1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fifth mean difference ; S3B2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the sixth mean difference ; S3B3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the seventh mean difference ; S3B4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the eighth mean. ; S3B5: Based on the calculation results of steps S3B1-S3B4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. ; S4: Perform a dot product operation on the first mean difference image and the second mean difference image to obtain a local mean difference minimum confidence image, and take the location of the maximum point in the local mean difference minimum confidence image as the location of the infrared weak target.

2. The infrared weak target detection method based on the minimum difference of local mean values ​​according to claim 1, characterized in that: Step S1 specifically includes: S11: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the first filtered image. ; S12: Utilizing dimensions of The mean filter is used to perform mean filtering on the original infrared image to obtain the second filtered image. ; S13: Subtract the first filtered image from the second filtered image to obtain the background-suppressed image. : ； in, These are the coordinates of the pixel's position on the image.

3. The infrared weak target detection method based on the minimum difference of local mean values ​​according to claim 1, characterized in that: In step S3, the specific steps for obtaining the first mean difference image are as follows: S3A1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the first mean difference : ； in, The mean filtering result centered on the target sub-window, obtained under the condition that the target sub-window is set as the mean filter and all elements of the other sub-windows are set to 0; The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 7 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 19 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 25 is set as the mean filter and all elements of the other sub-windows are set to 0. S3A2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference between the second means : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 11 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 12 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 14 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 15 is set as the mean filter and all elements of the other sub-windows are set to 0. S3A3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the third mean : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 21 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 17 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 5 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 9 is set as the mean filter and all elements of the other sub-windows are set to 0. S3A4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fourth mean difference : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 18 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 23 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 3 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 8 is set as the mean filter and all elements of the other sub-windows are set to 0. S3A5: Based on the calculation results of steps S3A1-S3A4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. : ； in, These are the coordinates of the pixel's position on the image.

4. The infrared weak target detection method based on the minimum difference of local mean values ​​according to claim 3, characterized in that: In step S3, the specific steps for obtaining the second mean difference image are as follows: S3B1: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the fifth mean difference : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 6 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 16 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 10 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 20 is set as the mean filter and all elements of the other sub-windows are set to 0. S3B2: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the sixth mean difference : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 22 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 4 is set as the mean filter and all elements of the other sub-windows are set to 0. S3B3: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the seventh mean difference : ； in, The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 24 is set as the mean filter and all elements of the other sub-windows are set to 0. The mean filtering result centered on the target sub-window is obtained under the condition that the sub-window numbered 2 is set as the mean filter and all elements of the other sub-windows are set to 0. S3B4: Based on mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Mean filtering results Calculate the difference of the eighth mean. : ； S3B5: Based on the calculation results of steps S3B1-S3B4, the minimum pixel value at each pixel position is taken using the following formula to obtain the second mean difference image. : ； in, These are the coordinates of the pixel's position on the image.

Images