Overwater highlight target identification method and system based on airborne infrared image

Abstract

The invention discloses a water highlight target identification method and system based on an airborne infrared image, and relates to the technical field of target identification. According to the technical key points, the method comprises the steps of obtaining an image containing an overwater highlight target; calculating a display cycle logarithm characteristic of the image; calculating a first segmentation threshold based on the image gray value, and performing first threshold segmentation on the display cycle logarithm feature map by using the first segmentation threshold; calculating the number of pixel points of each independent target in the image after the first threshold segmentation, and determining the independent target of which the number of the pixel points is smaller than a preset number threshold as a highlight target area; and calculating a second segmentation threshold based on the image gray value, performing second threshold segmentation on the image with the reserved highlight target area by using the second segmentation threshold, and determining the area greater than the second segmentation threshold as a highlight target. The method has stronger detail identification capability, can accurately distinguish the highlight target from the background, and effectively reduces the error identification rate.

Description

Technical Field

[0001] The present invention relates to the technical field of target recognition, and in particular to a method and system for recognizing a highlighted target on water based on airborne infrared images. Background Art

[0002] With the continuous development of infrared imaging technology and image feature extraction technology, their applications in the field of highlight target recognition are gradually merging. By applying image feature extraction technology to airborne infrared images, more accurate and efficient recognition of highlight targets can be achieved. At present, domestic and foreign scholars have carried out a large number of studies on highlight target recognition in airborne infrared images. These studies mainly focus on the optimization of feature extraction algorithms, the research of multi-feature fusion methods, and the application of machine learning and deep learning algorithms. Image feature extraction technology can be applied to the fields of maritime traffic monitoring, highlight target safety detection, fishery resource survey, etc., providing strong technical support for maritime safety and resource management. Summary of the invention

[0003] Therefore, the present invention proposes a method and system for identifying highlighted targets on water based on airborne infrared images, so as to meet the actual demand for accurately extracting highlighted targets on water in the field of highlighted target identification in aerial infrared images.

[0004] According to one aspect of the present invention, a method for identifying highlighted targets on water based on airborne infrared images is proposed, the method comprising: Acquire an image containing a highlighted target above water; Calculating a logarithmic feature of the image; the logarithmic feature of the image is defined as a logarithmic ratio of image saliency to the perimeter of the sliding window; Calculating a first segmentation threshold based on the gray value of the image, using the first segmentation threshold to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral image, marking an area greater than the first segmentation threshold as an independent target, and marking an area less than or equal to the first segmentation threshold as a background; Calculate the number of pixels of each independent target in the image after the first threshold segmentation, delete the independent targets whose number of pixels is greater than or equal to the preset number threshold, and determine the independent targets whose number of pixels is less than the preset number threshold as the highlighted target area; A second segmentation threshold is calculated based on the grayscale value of the image, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area greater than the second segmentation threshold as the highlighted target.

[0005] Furthermore, the calculating of the circumference logarithmic feature of the image includes: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ).

[0006] Furthermore, the calculation formula of the segmentation threshold in the sliding window is: T’ = μ + k σ' in, μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

[0007] Furthermore, the calculation formula of the first segmentation threshold is: T f = I min + m ×( I max – I mean ) in, I max is the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

[0008] Furthermore, the calculation formula of the second segmentation threshold is:

[0009] in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; pis a regulation constant greater than 1.

[0010] According to another aspect of the present invention, a water highlight target recognition system based on airborne infrared images is provided, the system comprising: An image acquisition module, configured to acquire an image containing a highlighted target above water; A feature calculation module configured to calculate a logarithmic feature of the image; the logarithmic feature of the image is defined as a logarithmic ratio of image saliency to a sliding window perimeter; A first segmentation module is configured to calculate a first segmentation threshold based on the image gray value, so as to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral display using the first segmentation threshold, mark the area greater than the first segmentation threshold as an independent target, and mark the area less than or equal to the first segmentation threshold as a background; A highlight region extraction module is configured to calculate the number of pixels of each independent target in the image after the first threshold segmentation, delete the independent targets whose number of pixels is greater than or equal to a preset number threshold, and determine the independent targets whose number of pixels is less than the preset number threshold as highlight target regions; The second segmentation module is configured to calculate a second segmentation threshold based on the image grayscale value, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area greater than the second segmentation threshold as the highlighted target.

[0011] Furthermore, the feature calculation module calculates the circumference logarithmic feature of the image, including: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ).

[0012] Furthermore, the calculation formula of the segmentation threshold in the sliding window in the feature calculation module is: T’ = μ + k σ' in, μ is the mean value of all pixel values ​​in the sliding window, σ'is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

[0013] Furthermore, the calculation formula of the first segmentation threshold in the first segmentation module is: T f = I min + m ×( I max – I mean ) in, I max is the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

[0014] Furthermore, the calculation formula of the second segmentation threshold in the second segmentation module is:

[0015] in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; p is a regulation constant greater than 1.

[0016] The beneficial technical effects of the present invention are: The present invention proposes a method and system for identifying highlighted targets on water based on airborne infrared images. First, an image containing highlighted targets on water is acquired; then, a logarithmic feature of the apparent circumference of the image is calculated; the logarithmic feature of the apparent circumference is defined as the logarithmic ratio of image saliency to the circumference of a sliding window; then, a first segmentation threshold is calculated based on the gray value of the image, so as to perform a first threshold segmentation on a feature map formed by the logarithmic feature of the apparent circumference using the first segmentation threshold, and an area greater than the first segmentation threshold is marked as an independent target, and an area less than or equal to the first segmentation threshold is marked as a background; then, the number of pixels of each independent target in the image after the first threshold segmentation is calculated, and independent targets with a number of pixels greater than or equal to a preset number threshold are deleted, and independent targets with a number of pixels less than the preset number threshold are determined as highlighted target areas; then, a second segmentation threshold is calculated based on the gray value of the image, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and an area greater than the second segmentation threshold is determined as a highlighted target area. The present invention has a stronger ability to identify details. By analyzing the thermal information in the infrared image, it can accurately distinguish the highlight target from the background and accurately extract the features of the highlight target in a complex background. The present invention can adapt to various different environments and noise conditions and effectively reduce the misidentification rate. The present invention achieves a high degree of automation and can still accurately identify even in a dynamically changing or low-contrast environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which: Figure 1 It is a flow chart of a method for identifying highlighted targets on water based on airborne infrared images according to an embodiment of the present invention.

[0018] Figure 2 This is an example image containing a highlighted target above water in an embodiment of the present invention.

[0019] Figure 3 1 is an example diagram of the logarithmic characteristic of the visible circumference calculated in an embodiment of the present invention.

[0020] Figure 4 This is an example feature diagram after the first threshold segmentation in an embodiment of the present invention.

[0021] Figure 5 This is an example diagram of a highlighted target area retained in an embodiment of the present invention.

[0022] Figure 6 It is an example image obtained by multiplying the image of the area where the highlighted target is located with the original image in an embodiment of the present invention.

[0023] Figure 7 is an example image of a highlighted target identified in an embodiment of the present invention.

[0024] Figure 8 It is a structural schematic diagram of a water highlight target recognition system based on airborne infrared imaging described in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0026] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software. In this article, it is to be understood that any number of elements in the drawings is for illustration and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0027] The embodiment of the present invention proposes a method for identifying highlighted targets on water based on airborne infrared images. Figure 1 As shown, the method includes: S1, acquiring an image containing a highlighted target above water; S2, calculating the logarithmic feature of the image; the logarithmic feature of the image is defined as the logarithmic ratio of the image saliency to the perimeter of the sliding window; S3, calculating a first segmentation threshold based on the image grayscale value, using the first segmentation threshold to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral image, marking an area greater than the first segmentation threshold as an independent target, and marking an area less than or equal to the first segmentation threshold as a background; S4, calculating the number of pixels of each independent target in the image after the first threshold segmentation, deleting the independent targets whose number of pixels is greater than or equal to the preset number threshold, and determining the independent targets whose number of pixels is less than the preset number threshold as the highlighted target area; S5. Calculate a second segmentation threshold based on the grayscale value of the image, use the second segmentation threshold to perform a second threshold segmentation on the image retaining the highlighted target area, and determine the area greater than the second segmentation threshold as the highlighted target.

[0028] In this embodiment, preferably, calculating the circumference logarithmic feature of the image in S2 includes: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ). The calculation formula of the segmentation threshold in the sliding window is: T’ = μ + k σ' in, μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

[0029] In this embodiment, preferably, the calculation formula of the first segmentation threshold in S3 is: T f = I min + m ×( I max – I mean ) in, I max is the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

[0030] In this embodiment, preferably, the calculation formula of the second segmentation threshold in S5 is:

[0031] in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; pis a regulation constant greater than 1.

[0032] The embodiments of the present invention are described in detail below.

[0033] First, in S1, an image containing a highlighted target above water is acquired.

[0034] According to an embodiment of the present invention, a drone carrying an infrared camera can be used to shoot a video containing a highlighted target, the water surface, and the sky close to the sea surface; and the video frame can be extracted to obtain an image containing a highlighted target on the water. Figure 2 shown.

[0035] Then, in S2, the circumference logarithmic feature of the image is calculated; the circumference logarithmic feature is defined as the logarithmic ratio of the image saliency to the circumference of the sliding window.

[0036] According to an embodiment of the present invention, the detailed calculation process of the logarithmic feature of the display period is as follows: 1) Set the sliding window size to 15×15; 2) Calculate the segmentation threshold within the sliding window: T’ = μ + k σ' in μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1. In this embodiment, k Take 2.3.

[0037] 3) Count the pixel values ​​in the sliding window that are greater than T' Number of pixels N T .

[0038] 4) Calculate the number of pixels around the sliding window N C : N C =4*(L-1) Where L is the number of pixels on the side of the sliding window.

[0039] 5) Calculate the image (i,j) Sliding window image features F ( i , j )for: F ( i , j )=log( N T ) / log( NC ) The resulting logarithmic characteristic diagram is as follows: Figure 3 shown.

[0040] Then, in S3, a first segmentation threshold is calculated based on the image grayscale value, and the feature map formed by the logarithmic feature of the peripheral image is subjected to a first threshold segmentation using the first segmentation threshold, and the area greater than the first segmentation threshold is marked as an independent target, and the area less than or equal to the first segmentation threshold is marked as the background.

[0041] According to an embodiment of the present invention, threshold segmentation is performed on the logarithmic feature map of the display period, and the segmentation threshold T f for: T f = I min + m ⋅( I max – I mean ) in, I max is the maximum grayscale value of the image, I min is the minimum grayscale value of the image, I mean is the image grayscale mean, m is a regulation constant in the range of [0,0.5]. The feature diagram after the first threshold segmentation is as follows: Figure 4 shown.

[0042] Then, in S4, the number of pixels of each independent target in the image after the first threshold segmentation is calculated, and the independent targets whose number of pixels is greater than or equal to the preset number threshold are deleted, and the independent targets whose number of pixels is less than the preset number threshold are determined as the highlighted target area.

[0043] According to an embodiment of the present invention, the number of pixels of each independent target in the computer-borne infrared image is T B , remove pixels greater than T B independent targets, such as coastlines, to obtain the area where the highlighted target is located. In this embodiment, set T B is 2000.

[0044] Examples of areas where highlighted targets are retained include: Figure 5 shown.

[0045] Furthermore, the image of the area where the highlighted target is located can be compared with the original image containing the highlighted target above water ( Figure 2) pixels, the resulting image is as follows Figure 6 shown.

[0046] Then, in S5, a second segmentation threshold is calculated based on the image grayscale value, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area greater than the second segmentation threshold as the highlighted target.

[0047] According to an embodiment of the present invention, the image is again segmented by a threshold value. T for:

[0048] in, and is the maximum and minimum value of all pixel (grayscale) values ​​in the image; σ is the standard deviation of the image pixel values, p is a regulation constant greater than 1. In this embodiment, p Take 2.5. After the second threshold segmentation, the highlighted target part in the image can be more accurately highlighted. The highlighted image example after recognition is as follows Figure 7 shown.

[0049] The present invention proposes a method for identifying highlighted targets on water based on airborne infrared images, which has a stronger ability to identify details. By analyzing the thermal information in the infrared image, the method can accurately distinguish the highlighted target from the background and accurately extract the features of the highlighted target under a complex background. In addition, the method can adapt to various different environments and noise conditions and effectively reduce the misrecognition rate. It also achieves a high degree of automation and can still accurately identify even in a dynamically changing or low-contrast environment.

[0050] Another embodiment of the present invention provides a water highlight target recognition system based on airborne infrared images, such as Figure 8 As shown, the system includes: An image acquisition module 810, configured to acquire an image containing a highlighted target above water; A feature calculation module 820, configured to calculate a logarithmic feature of the image; the logarithmic feature of the image is defined as a logarithmic ratio of image saliency to a sliding window perimeter; A first segmentation module 830 is configured to calculate a first segmentation threshold based on the image gray value, so as to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral image using the first segmentation threshold, mark the area greater than the first segmentation threshold as an independent target, and mark the area less than or equal to the first segmentation threshold as a background; A highlight region extraction module 840 is configured to calculate the number of pixels of each independent target in the image after the first threshold segmentation, delete the independent targets whose number of pixels is greater than or equal to a preset number threshold, and determine the independent targets whose number of pixels is less than the preset number threshold as highlight target regions; The second segmentation module 850 is configured to calculate a second segmentation threshold based on the image grayscale value, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area larger than the second segmentation threshold as the highlighted target.

[0051] In this embodiment, preferably, the feature calculation module 820 calculates the circumference logarithmic feature of the image, including: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ).

[0052] In this embodiment, preferably, the calculation formula of the segmentation threshold in the sliding window in the feature calculation module 820 is: T’ = μ + k σ' in, μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

[0053] In this embodiment, preferably, the calculation formula of the first segmentation threshold in the first segmentation module 830 is: T f = I min + m ×( I max – I mean ) in, I maxis the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

[0054] In this embodiment, preferably, the calculation formula of the second segmentation threshold in the second segmentation module 850 is:

[0055] in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; p is a regulation constant greater than 1.

[0056] It should be noted that the function of the water highlight target recognition system based on airborne infrared imaging described in this embodiment can be described by the aforementioned water highlight target recognition method based on airborne infrared imaging. For the parts not described in detail in the system embodiment, please refer to the above method embodiment.

[0057] It should be noted that, although several units, modules or submodules are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided into multiple modules to be embodied.

[0058] In addition, although the operations of the method of the present invention are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0059] Although the spirit and principle of the present invention have been described with reference to several specific embodiments, it should be understood that the present invention is not limited to the disclosed specific embodiments, and the division of various aspects does not mean that the features in these aspects cannot be combined to benefit, and such division is only for the convenience of expression. The present invention is intended to cover various modifications and equivalent arrangements included in the spirit and scope of the attached claims.

Claims

1. A method for identifying highlighted targets on water based on airborne infrared images, characterized in that: include: Acquire an image containing a highlighted target above water; Calculating the logarithmic feature of the image; The logarithmic feature of the saliency of the image is defined as the logarithmic ratio of the image saliency to the circumference of the sliding window; Calculating a first segmentation threshold based on the gray value of the image, using the first segmentation threshold to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral image, marking an area greater than the first segmentation threshold as an independent target, and marking an area less than or equal to the first segmentation threshold as a background; Calculate the number of pixels of each independent target in the image after the first threshold segmentation, delete the independent targets whose number of pixels is greater than or equal to the preset number threshold, and determine the independent targets whose number of pixels is less than the preset number threshold as the highlighted target area; A second segmentation threshold is calculated based on the grayscale value of the image, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area greater than the second segmentation threshold as the highlighted target.

2. The method for identifying highlighted targets on water based on airborne infrared images according to claim 1, characterized in that: The calculating the circumference logarithmic feature of the image comprises: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ).

3. The method for identifying highlighted targets on water based on airborne infrared images according to claim 2 is characterized in that: The calculation formula of the segmentation threshold in the sliding window is: T’ = μ + k σ' ； in, μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

4. The method for identifying highlighted targets on water based on airborne infrared images according to claim 1, characterized in that: The calculation formula of the first segmentation threshold is: T f = I min + m ×( I max – I mean )； in, I max is the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

5. The method for identifying highlighted targets on water based on airborne infrared images according to claim 1, characterized in that: The calculation formula of the second segmentation threshold is: ； in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; p is a regulation constant greater than 1.

6. A water highlight target recognition system based on airborne infrared images, characterized in that: include: An image acquisition module, configured to acquire an image containing a highlighted target above water; A feature calculation module configured to calculate a logarithmic feature of the image; the logarithmic feature of the image is defined as a logarithmic ratio of image saliency to a sliding window perimeter; A first segmentation module is configured to calculate a first segmentation threshold based on the image gray value, so as to perform a first threshold segmentation on the feature map formed by the logarithmic feature of the peripheral display using the first segmentation threshold, mark the area greater than the first segmentation threshold as an independent target, and mark the area less than or equal to the first segmentation threshold as a background; A highlight region extraction module is configured to calculate the number of pixels of each independent target in the image after the first threshold segmentation, delete the independent targets whose number of pixels is greater than or equal to a preset number threshold, and determine the independent targets whose number of pixels is less than the preset number threshold as highlight target regions; The second segmentation module is configured to calculate a second segmentation threshold based on the image grayscale value, so as to perform a second threshold segmentation on the image retaining the highlighted target area using the second segmentation threshold, and determine the area greater than the second segmentation threshold as the highlighted target.

7. The water highlight target recognition system based on airborne infrared imaging according to claim 6 is characterized in that: The feature calculation module calculates the circumference logarithmic feature of the image, including: Set the sliding window size and calculate the segmentation threshold within the sliding window based on the mean and standard deviation of all pixel values ​​within the sliding window; count the number of pixels within the sliding window whose pixel values ​​are greater than the segmentation threshold N T ; Calculate the number of pixels around the sliding window N C ; then the pixel (i,j) The corresponding logarithmic characteristics of the location are: F ( i , j )=log( N T ) / log( N C ).

8. The water highlight target recognition system based on airborne infrared imaging according to claim 7 is characterized in that: The calculation formula of the segmentation threshold in the sliding window in the feature calculation module is: T’ = μ + k σ' ； in, μ is the mean value of all pixel values ​​in the sliding window, σ' is the standard deviation of all pixel values ​​in the sliding window, k is a regulation constant greater than 1.

9. The water highlight target recognition system based on airborne infrared imaging according to claim 6 is characterized in that: The calculation formula of the first segmentation threshold in the first segmentation module is: T f = I min + m ×( I max – I mean )； in, I max is the maximum grayscale value of the logarithmic feature map of the visible period, I min is the minimum grayscale value of the logarithmic feature map of the visible period, I mean is the grayscale mean of the logarithmic feature map of the visible period, m is the adjustment constant.

10. The water highlight target recognition system based on airborne infrared imaging according to claim 6, characterized in that: The calculation formula of the second segmentation threshold in the second segmentation module is: ； in, and To retain the maximum and minimum grayscale values ​​of the image in the highlighted target area; σ The standard deviation of all gray values ​​of the image to retain the highlighted target area; p is a regulation constant greater than 1.

Images