Data processing method for optoelectronic pod

Abstract

The present invention discloses a data processing method for an optoelectronic pod, which belongs to the field of image processing technology, and includes the following steps: S1, dividing the target image collected by the optoelectronic pod to obtain four quadrant areas; S2, calculating the correction value of the target image, determining the area mark of each quadrant area, and generating an outer edge pixel point set of the target image; S3, processing the outer edge pixel point set to generate a clear target image of the optoelectronic pod. The present invention fully considers the color value distribution of the pixel points in each quadrant area, can adapt to the influence of different lighting conditions and noise levels on the image quality, and accurately realizes edge detection.

Description

Technical Field

[0001] The invention belongs to the technical field of image processing, and in particular relates to a data processing method for an optoelectronic pod. Background Art

[0002] The optoelectronic pod is an important part of optoelectronic reconnaissance technology equipment. It can carry infrared and visible light image sensors to complete target detection and tracking tasks. As the optoelectronic pod reconnaissance image data becomes increasingly abundant, the image resolution continues to improve, the scenes in the images become more and more complex, and the scales of objects become more and more diverse, which makes it more difficult to segment the boundaries of objects in optoelectronic pod reconnaissance images. At the same time, due to the diversity and complexity of the ground object information in the optoelectronic pod reconnaissance image, the noise, illumination changes, camera distortion and other factors that may exist in the image acquisition process, the edge of the image often appears blurred, distorted or irregular. Therefore, it has become an urgent problem to effectively process the edges of the optoelectronic pod image data to improve the clarity and accuracy of the image. Summary of the invention

[0003] In order to solve the above problems, the present invention proposes a data processing method for an optoelectronic pod.

[0004] The technical solution of the present invention is: a data processing method of an optoelectronic pod comprises the following steps:

[0005] S1. Divide the target image collected by the optoelectronic pod into four quadrants;

[0006] S2, calculating the correction value of the target image, determining the region mark of each quadrant region, and generating a set of outer edge pixel points of the target image;

[0007] S3. Process the set of outer edge pixels to generate a clear target image of the optoelectronic pod.

[0008] Furthermore, S2 includes the following sub-steps:

[0009] S21, calculating the weighted regression value of each pixel point based on the quadrant area where each pixel point is located;

[0010] S22, arranging the weighted regression values ​​of all pixels in descending order, and calculating the correction value of the target image;

[0011] S23, determining a region mark in each quadrant region according to the correction value of the target image;

[0012] S24, connecting the region mark of the first quadrant area, the region mark of the second quadrant area, the region mark of the third quadrant area and the region mark of the fourth quadrant area in sequence to form a closed area, and taking the pixel points in the target image except the closed area as the outer edge pixel point set.

[0013] The beneficial effects of the above further scheme are: in the present invention, by considering the quadrant area where the pixel point is located and combining its position information to calculate the weight regression value, the importance of the pixel point in the image can be more accurately reflected; the correction value is calculated based on the weight regression value of all pixels, which can more comprehensively reflect the overall characteristics of the image; by forming a closed area, the edge information in the image can be clearly identified.

[0014] Furthermore, in S21, the weighted regression value Z of the i-th pixel is i The calculation formula is:

[0015] ; In the formula, R i Represents the red component value of the i-th pixel, G i Indicates the green component value of the i-th pixel, B i Represents the blue component value of the i-th pixel, R i_max Indicates the maximum red component value of the quadrant where the i-th pixel is located, R i_min Indicates the minimum red component value of the quadrant where the i-th pixel is located, G i_max Indicates the maximum green component value of the quadrant where the i-th pixel is located, G i_min Indicates the minimum green component value of the quadrant where the i-th pixel is located, B i_max Indicates the maximum blue component value of the quadrant where the i-th pixel is located, B i_min Indicates the minimum blue component value of the quadrant where the i-th pixel is located. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum blue component value in the quadrant. Represents the Euclidean distance between the i-th pixel and the pixel with the minimum blue component value in the quadrant.

[0016] Furthermore, in S22, the calculation formula of the correction value X of the target image is:

[0017] ; In the formula, Z1 represents the regression value of the first pixel, Z Irepresents the regression value of the I-th pixel, max(·) represents the maximum function, min(·) represents the minimum function, and median(·) represents the median function.

[0018] Further, S23 includes the following sub-steps:

[0019] S231, generating an attention matrix according to the correction value of the target image;

[0020] S232. Based on the attention matrix, calculate the attention degree of each pixel in the quadrant area, and use the pixel corresponding to the maximum attention degree in each quadrant area as the area marker of the quadrant area.

[0021] The beneficial effect of the above further scheme is that in the present invention, by generating an attention matrix, the key areas in the image can be quickly identified, reflecting the significance of different areas in the image, and the image content can be better understood. By calculating the attention of each pixel in the quadrant area and selecting the pixel corresponding to the maximum attention as the area marker, the key area in the image can be accurately located.

[0022] Furthermore, in S231, the expression of the attention matrix N is:

[0023] , ; In the formula, n represents the initial attention matrix, ln(·) represents the logarithmic function, and e represents the exponent. Represents the mean of the weighted regression values ​​of all pixels in the first quadrant, Represents the mean of the weighted regression values ​​of all pixels in the second quadrant, Represents the mean of the weighted regression values ​​of all pixels in the third quadrant, represents the mean value of the weighted regression values ​​of all pixels in the fourth quadrant, and X represents the correction value of the target image.

[0024] Further, in S232, the calculation formula of the attention degree φ of the pixel point in the quadrant area is: ; Where q represents the maximum singular value of the attention matrix, and P represents the RGB value of the pixel.

[0025] Furthermore, in S3, a smoothing filter is used to smooth the set of outer edge pixels.

[0026] The beneficial effects of the present invention are as follows: the present invention divides the target image into four quadrant areas, determines the set of outer edge pixel points based on the area marks of each quadrant area, and can more accurately identify the edge of the image. By comprehensively considering the information of the four quadrants, the misjudgment rate of edge detection can be reduced and the integrity and continuity of the edge can be improved. The present invention fully considers the color value distribution of pixel points in each quadrant area, can adapt to the influence of different lighting conditions and noise levels and other factors on image quality, and accurately realizes edge detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 The figure is a flow chart of the data processing method of the optoelectronic pod. DETAILED DESCRIPTION

[0028] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0029] like Figure 1 As shown, the present invention provides a data processing method for an optoelectronic pod, comprising the following steps:

[0030] S1. Divide the target image collected by the optoelectronic pod into four quadrants;

[0031] S2, calculating the correction value of the target image, determining the region mark of each quadrant region, and generating a set of outer edge pixel points of the target image;

[0032] S3. Process the set of outer edge pixels to generate a clear target image of the optoelectronic pod.

[0033] In this embodiment of the present invention, S2 includes the following sub-steps:

[0034] S21, calculating the weighted regression value of each pixel point based on the quadrant area where each pixel point is located;

[0035] S22, arranging the weighted regression values ​​of all pixels in descending order, and calculating the correction value of the target image;

[0036] S23, determining a region mark in each quadrant region according to the correction value of the target image;

[0037] S24, connecting the region mark of the first quadrant area, the region mark of the second quadrant area, the region mark of the third quadrant area and the region mark of the fourth quadrant area in sequence to form a closed area, and taking the pixel points in the target image except the closed area as the outer edge pixel point set.

[0038] In the present invention, by considering the quadrant area where the pixel point is located and combining its position information to calculate the weighted regression value, the importance of the pixel point in the image can be more accurately reflected; the correction value is calculated based on the weighted regression value of all pixels, which can more comprehensively reflect the overall characteristics of the image; by forming a closed area, the edge information in the image can be clearly identified.

[0039] In the embodiment of the present invention, in S21, the weighted regression value Z of the i-th pixel point i The calculation formula is:

[0040] ; In the formula, R i Represents the red component value of the i-th pixel, G i Indicates the green component value of the i-th pixel, B i Represents the blue component value of the i-th pixel, R i_max Indicates the maximum red component value of the quadrant where the i-th pixel is located, R i_min Indicates the minimum red component value of the quadrant where the i-th pixel is located, G i_max Indicates the maximum green component value of the quadrant where the i-th pixel is located, G i_min Indicates the minimum green component value of the quadrant where the i-th pixel is located, B i_max Indicates the maximum blue component value of the quadrant where the i-th pixel is located, B i_min Indicates the minimum blue component value of the quadrant where the i-th pixel is located. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum blue component value in the quadrant. Represents the Euclidean distance between the i-th pixel and the pixel with the minimum blue component value in the quadrant.

[0041] In the embodiment of the present invention, in S22, the calculation formula of the correction value X of the target image is:

[0042] ; In the formula, Z1 represents the regression value of the first pixel, Z Irepresents the regression value of the I-th pixel, max(·) represents the maximum function, min(·) represents the minimum function, and median(·) represents the median function.

[0043] In this embodiment of the present invention, S23 includes the following sub-steps:

[0044] S231, generating an attention matrix according to the correction value of the target image;

[0045] S232. Based on the attention matrix, calculate the attention degree of each pixel in the quadrant area, and use the pixel corresponding to the maximum attention degree in each quadrant area as the area marker of the quadrant area.

[0046] In the present invention, by generating an attention matrix, the key areas in the image can be quickly identified, which reflects the significance of different areas in the image and can better understand the image content. By calculating the attention of each pixel in the quadrant area and selecting the pixel corresponding to the maximum attention as the area marker, the key areas in the image can be accurately located.

[0047] In the embodiment of the present invention, in S231, it is noted that the expression of the matrix N is:

[0048] , ; In the formula, n represents the initial attention matrix, ln(·) represents the logarithmic function, and e represents the exponent. Represents the mean of the weighted regression values ​​of all pixels in the first quadrant, Represents the mean of the weighted regression values ​​of all pixels in the second quadrant, Represents the mean of the weighted regression values ​​of all pixels in the third quadrant, represents the mean value of the weighted regression values ​​of all pixels in the fourth quadrant, and X represents the correction value of the target image.

[0049] In the embodiment of the present invention, in S232, the calculation formula of the attention degree φ of the pixel point in the quadrant area is: ; Where q represents the maximum singular value of the attention matrix, and P represents the RGB value of the pixel.

[0050] In the embodiment of the present invention, in S3, a smoothing filter is used to perform smoothing processing on the outer edge pixel point set.

[0051] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A data processing method for an optoelectronic pod, characterized in that: The following steps are involved: S1. Divide the target image collected by the optoelectronic pod into four quadrants; S2, calculating the correction value of the target image, determining the region mark of each quadrant region, and generating a set of outer edge pixel points of the target image; S3, processing the outer edge pixel point set to generate a clear target image of the optoelectronic pod; The S2 comprises the following sub-steps: S21, calculating the weighted regression value of each pixel point based on the quadrant area where each pixel point is located; S22, arranging the weighted regression values ​​of all pixels in descending order, and calculating the correction value of the target image; S23, determining a region mark in each quadrant region according to the correction value of the target image; S24, sequentially connecting the region mark of the first quadrant region, the region mark of the second quadrant region, the region mark of the third quadrant region, and the region mark of the fourth quadrant region to form a closed region, and taking the pixel points in the target image except the closed region as an outer edge pixel point set; In S21, the weighted regression value Z of the i-th pixel point i The calculation formula is: ; In the formula, R i Represents the red component value of the i-th pixel, G i Indicates the green component value of the i-th pixel, B i Represents the blue component value of the i-th pixel, R i_max Indicates the maximum red component value of the quadrant where the i-th pixel is located, R i_min Indicates the minimum red component value of the quadrant where the i-th pixel is located, G i_max Indicates the maximum green component value of the quadrant where the i-th pixel is located, G i_min Indicates the minimum green component value of the quadrant where the i-th pixel is located, B i_max Indicates the maximum blue component value of the quadrant where the i-th pixel is located, B i_min Indicates the minimum blue component value of the quadrant where the i-th pixel is located. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum red component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum green component value in the quadrant. It represents the Euclidean distance between the i-th pixel and the pixel corresponding to the maximum blue component value in the quadrant. Represents the Euclidean distance between the i-th pixel and the pixel corresponding to the minimum blue component value in the quadrant; In S22, the calculation formula of the correction value X of the target image is: ; In the formula, Z1 represents the regression value of the first pixel, Z I represents the regression value of the I-th pixel, max(·) represents the maximum function, min(·) represents the minimum function, and median(·) represents the median function; The S23 comprises the following sub-steps: S231, generating an attention matrix according to the correction value of the target image; S232, based on the attention matrix, calculating the attention degree of each pixel in the quadrant area, and taking the pixel corresponding to the maximum attention degree in each quadrant area as the area mark of the quadrant area; In S231, note that the expression of the matrix N is: , ; In the formula, n represents the initial attention matrix, ln(·) represents the logarithmic function, and e represents the exponent. Represents the mean of the weighted regression values ​​of all pixels in the first quadrant, Represents the mean of the weighted regression values ​​of all pixels in the second quadrant, represents the mean of the weighted regression values ​​of all pixels in the third quadrant, represents the mean value of the weighted regression values ​​of all pixels in the fourth quadrant, and X represents the correction value of the target image; In S232, the calculation formula of the attention degree φ of the pixel point in the quadrant area is: ; In the formula, q represents the maximum singular value of the attention matrix, and P represents the RGB value of the pixel; In S3, a smoothing filter is used to smooth the outer edge pixel point set.

Images