A method for processing UAV remote sensing mapping images

Abstract

The present invention discloses a method for processing unmanned aerial vehicle remote sensing mapping images, which relates to the technical field of image processing, and specifically includes: Step S1: Collecting image data of a target area to obtain a set of mapping images to be processed; Step S2: Performing gray mapping on the set of mapping images to be processed to obtain an initial gray image set, and performing denoising processing on the initial gray image set through median filtering with the introduction of a window weight factor to obtain denoised gray images; Step S3: Performing calculation processing on the denoised gray images through inverse gray transformation and color channel restoration algorithms to obtain color-enhanced mapping images; Step S4: Performing deblurring processing on the color-enhanced mapping images, calculating the gray main value area representative, fixed segmentation threshold, and environmental light interference amount based on the denoised gray images and color-enhanced mapping images, and obtaining deblurred mapping images through dynamic segmentation threshold calculation; Step S5: Performing registration and stitching processing on the deblurred remote sensing mapping images to obtain a panoramic registration image.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method for processing unmanned aerial vehicle (UAV) remote sensing mapping images. Background Art

[0002] In recent years, with the rapid development of sensor technology, communication technology, and automatic control technology, the performance of UAVs has been continuously improved, and they can now undertake a large amount of remote sensing mapping work. Compared with manual mapping and satellite mapping, UAVs have prominent advantages such as low cost, high speed, and flexible data acquisition. With the increasing application of UAV remote sensing mapping, the optimized development of UAV remote sensing mapping image processing is a promising research direction.

[0003] Currently, Chinese Patent Application No. CN202411472333.4 discloses a method for processing UAV remote sensing mapping images. The method includes: S1. Collecting remote sensing mapping images captured by a UAV, generating a step function for each row of the remote sensing mapping images, and determining the corresponding change curve; S2. Generating a sharpening kernel using the change curve; S3. Enhancing the pixel points of the remote sensing mapping images using the sharpening kernel. The present invention can make the enhanced remote sensing mapping images look clearer by enhancing the edges and contours in the remote sensing mapping images, facilitating further analysis and processing of the images by users. However, this application sharpens the image contours through a kernel function and lacks enhancement processing for the color direction of the images. Summary of the Invention

[0004] The technical problems solved by the present invention are as follows: When denoising using traditional methods, all images are uniformly processed without adaptive adjustment. At the same time, when enhancing remote sensing mapping images, mainly edge and contour sharpening processing is carried out, lacking enhancement processing for the color direction.

[0005] To solve the above technical problems, the present invention provides the following technical solutions:

[0006] A method for processing UAV remote sensing mapping images, including:

[0007] Step S1: Collecting image data of the target area to obtain a set of to-be-processed mapping images;

[0008] Step S2: Performing gray mapping on the set of to-be-processed mapping images to obtain an initial set of gray images, and denoising the initial set of gray images by median filtering with the introduction of a window weight factor to obtain denoised gray images;

[0009] Step S3: Calculating and processing the denoised gray images through inverse gray transformation and color channel restoration algorithms to obtain color-enhanced mapping images;

[0010] Step S4: Perform deblurring on the color-enhanced mapping image, calculate the grayscale main value area representative, fixed segmentation threshold, and environmental light interference amount based on the denoised grayscale image and the color-enhanced mapping image, and obtain the deblurred mapping image through dynamic segmentation threshold calculation;

[0011] Step S5: Perform registration and stitching on the deblurred remote sensing mapping image to obtain a panoramic registration image.

[0012] As a preferred solution of the method for processing unmanned aerial vehicle remote sensing mapping images according to the present invention, wherein: control the unmanned aerial vehicle to fly based on the planned inspection route, and collect image data of the target mapping area through the remote sensing optical sensor on the unmanned aerial vehicle platform to obtain a set of to-be-processed mapping images.

[0013] As a preferred solution of the method for processing unmanned aerial vehicle remote sensing mapping images according to the present invention, wherein: perform grayscale mapping on each image in the set of to-be-processed mapping images to obtain an initial grayscale image set, perform partition processing on the initial grayscale image set through a sliding window with a preset size and step length to obtain window sub-images, calculate the window grayscale median corresponding to each window sub-image through sorting calculation, and calculate a window weight factor based on the window grayscale median corresponding to each window sub-image.

[0014] As a preferred solution of the method for processing unmanned aerial vehicle remote sensing mapping images according to the present invention, wherein: the processing logic for calculating the window weight factor includes:

[0015] Calculate the grayscale average value of all pixel points of each window sub-image through a single averaging calculation, calculate the absolute value of the difference between the grayscale average value corresponding to each window sub-image and the window grayscale median respectively, perform a secondary averaging calculation on the discrete data corresponding to each window sub-image to obtain window discrete data, obtain the window weight factor based on the pixel point discrete data and the window discrete data, and perform weighted calculation on the corresponding sliding window based on the window weight factor to obtain a denoised grayscale image, and its calculation expression is:

[0016] ;

[0017] ;

[0018] ;

[0019] wherein, x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, G(x, y) represents the grayscale value of the pixel point at the coordinate (x, y), represents the grayscale average value of the window sub-image, DAV(x, y) represents the pixel point discrete data, AAV(x, y) represents the window discrete data, and N represents the total number of pixel points of the window sub-image. w represents the window weight factor, c represents the window weight factor constant, and max{·} represents the maximum value function.

[0020] As a preferred solution of a method for processing UAV remote sensing mapping images according to the present invention, wherein: the denoised grayscale image is subjected to inverse grayscale transformation to obtain a denoised original color mapping image, and the denoised original color mapping image is calculated and processed by an improved color channel restoration algorithm to obtain a color-enhanced mapping image.

[0021] As a preferred solution of a method for processing UAV remote sensing mapping images according to the present invention, wherein: the processing logic of the improved color channel restoration algorithm includes:

[0022] Based on a preset first color depth value, the denoised original color mapping image is subjected to color quantization reduction processing to obtain a low-color-depth image, the low-color-depth image is subjected to color enhancement calculation to obtain a transition enhancement image, and based on a preset second color depth value, the transition enhancement image is subjected to color quantization improvement processing to obtain a color-enhanced mapping image, and its calculation expression is:

[0023] ;

[0024] ;

[0025] wherein, x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, PT(x, y) represents the transition enhancement image, PI(x, y) represents the low-color-depth image, e represents the natural constant, MF(x, y) represents the color enhancement function, represents the mapping constant, represents the gain constant, BC(x, y) represents the color band in the low-color-depth image, and M represents the total number of color bands in the low-color-depth image.

[0026] As a preferred solution of a method for processing UAV remote sensing mapping images according to the present invention, wherein: the color-enhanced mapping image is subjected to deblurring processing to obtain a deblurred mapping image, and the processing logic includes:

[0027] Compare the values of the R channel, G channel, and B channel of each pixel point in the color-enhanced mapping image and take the minimum value to obtain the dark channel value, traverse all pixel points in the color-enhanced mapping image to obtain the dark channel mapping image, sort the pixel point values in the dark channel mapping image from high to low, and select the largest U% pixel values in the dark channel mapping image based on a preset ambient light ratio coefficient U% and sum them to obtain the ambient light interference amount;

[0028] A grayscale histogram is calculated based on the denoised grayscale image. The grayscale probability distribution is obtained by linearly mapping the grayscale histogram through the Min-Max normalization algorithm. The cross-correlation coefficient between the grayscale probability distribution and the standard normal distribution function is calculated. The grayscale probability distribution is threshold-screened through a preset cross-correlation parameter, and the grayscale values greater than or equal to the preset cross-correlation parameter in the grayscale probability distribution are selected as the grayscale main value area. The median value processing is performed on the grayscale main value area to obtain the representative of the grayscale main value area;

[0029] The fixed segmentation threshold is calculated for the grayscale histogram through Otsu's algorithm, and the dynamic segmentation threshold is calculated based on the representative of the grayscale main value area, the fixed segmentation threshold, and the environmental light interference amount;

[0030] Its calculation expression is:

[0031] ;

[0032] Among them, DST represents the dynamic segmentation threshold, max(·) represents the maximum value function, ALI represents the environmental light interference amount, PGR represents the representative of the grayscale main value area, and BST represents the fixed segmentation threshold.

[0033] As a preferred solution of the method for processing unmanned aerial vehicle remote sensing mapping images described in the present invention, wherein: the denoised grayscale image is segmented based on the dynamic segmentation threshold to obtain the first blurred area image and the first non-blurred area image, and the dark channel mapping image is mask-segmented based on the pixel point coordinates of the first blurred area image and the first non-blurred area image to obtain the second blurred area image and the second non-blurred area image;

[0034] The deblurring dynamic coefficient is calculated based on the first blurred area image and the dynamic segmentation threshold, and the blurred area enhancement image is obtained by multiplying the deblurring dynamic coefficient and the second blurred area image;

[0035] The deblurred dark channel image is obtained by performing image stitching processing on the blurred area enhancement image and the second non-blurred area image, and the deblurred remote sensing mapping image is obtained by performing restoration processing based on the atmospheric light scattering model and the deblurred dark channel image;

[0036] The calculation expression of the deblurring dynamic coefficient is:

[0037] ;

[0038] Among them, x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, DEV(x,y) represents the deblurring dynamic coefficient, represents the limited constant, FGR(x,y) represents the grayscale value of the pixel point of the first blurred area image, and DST represents the dynamic segmentation threshold.

[0039] As a preferred solution of a method for processing UAV remote sensing mapping images according to the present invention, the following steps are included: a reference image and an image to be registered are selected based on the deblurred remote sensing mapping image, and the reference image and the image to be registered are registered and stitched to obtain a panoramic registered image.

[0040] As a preferred solution of a method for processing UAV remote sensing mapping images according to the present invention, the processing logic for registering the image to be registered to obtain a panoramic registered image includes:

[0041] The Harris matrix is used to calculate the first corner feature vector group and the second corner feature vector group for the reference image and the image to be registered respectively;

[0042] The reference image and the image to be registered are subjected to FFT calculation through the fast Fourier transform to obtain the first phase angle data set and the second phase angle data. The frequency domain phase data in the first phase angle data set is extracted and concatenated with the vectors in the corresponding first corner feature vector group to obtain the first descriptor vector set. The frequency domain phase data in the second phase angle data set is extracted and concatenated with the vectors in the corresponding second corner feature vector group to obtain the second descriptor vector set;

[0043] Based on the first descriptor vector set and the second descriptor vector set, a primary registration determination is performed. The Euclidean distance values between the elements in the first descriptor vector set and the second descriptor vector set are calculated. When the Euclidean distance is greater than or equal to the preset first registration threshold, it is determined as pre-pairing of feature points;

[0044] Based on the pre-pairing of feature points, a secondary registration determination is performed. The number of pixel points forming pre-pairing of feature points in the image to be registered is statistically calculated through a registration sliding window of Q*Q size. When the number of pixel points forming pre-pairing of feature points in the image to be registered is greater than or equal to the preset second registration threshold, it is retained to obtain the final pairing of feature points;

[0045] Based on the final pairing of feature points of the reference image and the image to be registered, the registration calculation is performed through the affine transformation model to obtain the panoramic registered image.

[0046] The beneficial effects of the present invention are as follows: By introducing a window weight factor, it can be adaptively adjusted according to the gray distribution of pixels within the window, which is beneficial to improving the pertinence and accuracy of denoising. An improved color channel restoration algorithm is established by using color depth transformation, and active color quantization reduction processing is performed to form color bands in the image, which is beneficial to highlighting the color distribution information of the image and enhancing the color expressiveness and image resolution of the remote sensing mapping image. The dynamic segmentation threshold is calculated by integrating the dark channel and the gray histogram, optimizing the traditional atmospheric light deblurring model. Feature descriptors are formed through the Harris matrix and FFT, and two-stage threshold determination is performed, which is beneficial to accurately registering to obtain a panoramic image. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 Schematic diagram of the basic process of a method for processing unmanned aerial vehicle remote sensing mapping images provided by an embodiment of the present invention. Specific embodiments

[0048] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments.

[0049] Refer to Figure 1 , an embodiment of the present invention provides a method for processing unmanned aerial vehicle remote sensing mapping images, including:

[0050] Step S1: Collect image data of the target area to obtain a set of to-be-processed mapping images;

[0051] Step S2: Perform gray mapping on the set of to-be-processed mapping images to obtain an initial gray image set, and perform denoising processing on the initial gray image set through median filtering introducing a window weight factor to obtain denoised gray images;

[0052] Step S3: Perform calculation processing on the denoised gray images through inverse gray transformation and color channel restoration algorithms to obtain color-enhanced mapping images;

[0053] Step S4: Perform deblurring processing on the color-enhanced mapping images, calculate the gray main value area representative, fixed segmentation threshold, and environmental light interference amount based on the denoised gray images and color-enhanced mapping images, and obtain deblurred mapping images through dynamic segmentation threshold calculation;

[0054] Step S5: Perform registration and stitching processing on the deblurred remote sensing mapping images to obtain a panoramic registration image.

[0055] In this embodiment, the unmanned aerial vehicle is controlled to fly based on a planned inspection route, and image data of the target mapping area is collected through a remote sensing optical sensor on the unmanned aerial vehicle platform to obtain a set of to-be-processed mapping images.

[0056] In this embodiment, each image in the set of to-be-processed mapping images is respectively subjected to gray mapping to obtain an initial gray image set, the initial gray image set is partitioned through a sliding window with a preset size and step length to obtain window sub-images, the window gray median corresponding to each window sub-image is obtained through sorting calculation, and a window weight factor is calculated based on the window gray median corresponding to each window sub-image.

[0057] In this embodiment, the processing logic for calculating the window weight factor includes:

[0058] The grayscale average value of all pixel points in each window sub-image is calculated by taking an average once. The absolute value is obtained by subtracting the grayscale average value corresponding to each window sub-image from the median grayscale value of the window, resulting in pixel discrete data. The discrete data corresponding to each window sub-image is averaged again to obtain window discrete data. Based on the pixel discrete data and the window discrete data, a window weight factor is obtained. The sliding window corresponding to the window weight factor is weighted to obtain a denoised grayscale image, and its calculation expression is:

[0059] ;

[0060] ;

[0061] ;

[0062] where x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, G(x, y) represents the grayscale value of the pixel point at the coordinate (x, y), represents the grayscale average value of the window sub-image, DAV(x, y) represents the pixel discrete data, AAV(x, y) represents the window discrete data, N represents the total number of pixel points in the window sub-image, represents the window weight factor, c represents the window weight factor constant, and max{·} represents the maximum value function.

[0063] Among them, the collected remote sensing mapping image has system interferences such as electromagnetic noise. Grayscale processing is beneficial for single-channel calculation, improves calculation efficiency, and provides a grayscale image for subsequent deblurring calculation. Innovatively introducing a window weight factor can be adaptively adjusted according to the grayscale distribution of pixels within the window, which is beneficial for improving the pertinence and accuracy of denoising.

[0064] In this embodiment, an inverse grayscale transformation is performed on the denoised grayscale image to obtain a denoised original color mapping image, and a color enhancement mapping image is obtained by performing calculation processing on the denoised original color mapping image through an improved color channel restoration algorithm.

[0065] In this embodiment, the processing logic of the improved color channel restoration algorithm includes:

[0066] Performing color quantization reduction processing on the denoised original color mapping image based on a preset first color depth value to obtain a low-color-depth image, performing color enhancement calculation on the low-color-depth image to obtain a transitional enhancement image, and performing color quantization improvement processing on the transitional enhancement image based on a preset second color depth value to obtain a color enhancement mapping image, and its calculation expression is:

[0067] ;

[0068] ;

[0069] Among them, x represents the abscissa of the pixel, y represents the ordinate of the pixel, PT(x, y) represents the transition enhanced image, PI(x, y) represents the low-color-depth image, e represents the natural constant, MF(x, y) represents the color enhancement function, represents the mapping constant, represents the gain constant, BC(x, y) represents the color band in the low-color-depth image, and M represents the total number of color bands in the low-color-depth image.

[0070] Among them, actively performing color quantization reduction processing to form color bands in the image is beneficial to highlighting the color distribution information of the image, enhancing color contrast, and enhancing the image recognition of remote sensing mapping.

[0071] In this embodiment, the color-enhanced mapping image is de-blurred to obtain a de-blurred mapping image, and the processing logic includes:

[0072] Compare the values of the R channel, G channel, and B channel of each pixel in the color-enhanced mapping image and take the minimum value to obtain the dark channel value. Traverse all pixels in the color-enhanced mapping image to obtain the dark channel mapping image, sort the pixel values in the dark channel mapping image from high to low, and select the largest U% pixel values in the dark channel mapping image based on the preset environmental illumination ratio coefficient U% and sum them to obtain the environmental illumination interference amount;

[0073] Calculate the gray histogram based on the denoised gray image, linearly map the gray histogram through the Min-Max normalization algorithm to obtain the gray probability distribution, calculate the cross-correlation coefficient between the gray probability distribution and the standard normal distribution function, perform threshold screening on the gray probability distribution through the preset cross-correlation parameter, select the gray values greater than or equal to the preset cross-correlation parameter in the gray probability distribution as the gray main value area, and perform median value processing on the gray main value area to obtain the gray main value area representative;

[0074] Calculate the fixed segmentation threshold through Otsu's algorithm for the gray histogram, and calculate the dynamic segmentation threshold based on the gray main value area representative, the fixed segmentation threshold, and the environmental illumination interference amount;

[0075] Its calculation expression is:

[0076] ;

[0077] Among them, DST represents the dynamic segmentation threshold, max(·) represents the maximum value function, ALI represents the environmental illumination interference amount, PGR represents the gray main value area representative, and BST represents the fixed segmentation threshold.

[0078] In this embodiment, the denoised grayscale image is segmented based on a dynamic segmentation threshold to obtain a first blurred region image and a first non-blurred region image. Then, based on the pixel coordinates of the first blurred region image and the first non-blurred region image, the dark channel mapping image is subjected to mask segmentation processing to obtain a second blurred region image and a second non-blurred region image;

[0079] Based on the first blurred region image and the dynamic segmentation threshold, a deblurring dynamic coefficient is calculated. The deblurring dynamic coefficient and the second blurred region image are multiplied to obtain a blurred region enhancement image;

[0080] Based on the blurred region enhancement image and the second non-blurred region image, image stitching processing is performed to obtain a deblurred dark channel image. Based on the atmospheric light scattering model and the deblurred dark channel image, restoration processing is performed to obtain a deblurred remote sensing mapping image;

[0081] The calculation expression of the deblurring dynamic coefficient is:

[0082] ;

[0083] Where x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, DEV(x, y) represents the deblurring dynamic coefficient, represents a defined constant, FGR(x, y) represents the gray value of the pixel point in the first blurred region image, and DST represents the dynamic segmentation threshold.

[0084] In this embodiment, a reference image and a to-be-registered image are selected based on the deblurred remote sensing mapping image, and registration and stitching processing are performed on the reference image and the to-be-registered image to obtain a panoramic registration image.

[0085] Among them, the area of the target region of remote sensing mapping is often large. Different from satellites, the single-frame images collected by drones have a small area, and there is an overlapping phenomenon among the deblurred remote sensing mapping images. Multiple images need to be merged to obtain a panoramic image of the target region. Registration is beneficial to obtain an accurate panoramic large image. In this example, L non-overlapping deblurred remote sensing mapping images are used as the reference images, and the remaining (total number - L) overlapping deblurred remote sensing mapping images are used as the to-be-registered images.

[0086] In this embodiment, the processing logic for performing registration processing on the to-be-registered image to obtain a panoramic registration image includes:

[0087] The Harris matrix is used to calculate the first corner feature vector group and the second corner feature vector group for the reference image and the to-be-registered image respectively;

[0088] Perform FFT calculations on the reference image and the image to be registered through the fast Fourier transform to obtain the first phase angle data set and the second phase angle data. Extract the frequency domain phase data in the first phase angle data set and concatenate it with the vectors in the corresponding first corner feature vector group to obtain the first descriptor vector set. Extract the frequency domain phase data in the second phase angle data set and concatenate it with the vectors in the corresponding second corner feature vector group to obtain the second descriptor vector set;

[0089] Based on the first descriptor vector set and the second descriptor vector set, perform a primary registration determination. Calculate the Euclidean distance values between the elements in the first descriptor vector set and the second descriptor vector set. When the Euclidean distance is greater than or equal to the preset first registration threshold, it is determined as a pre-matching of feature points;

[0090] Based on the pre-matching of feature points, perform a secondary registration determination. Statistically count the number of pixel points that form pre-matching of feature points in the image to be registered through a registration sliding window of size Q*Q. When the number of pixel points that form pre-matching of feature points in the image to be registered is greater than or equal to the preset second registration threshold, retain them to obtain the final matching of feature points;

[0091] Based on the final matching of the feature points of the reference image and the image to be registered, perform registration calculations through the affine transformation model to obtain the panoramic registration image.

[0092] Among them, combine the Harris matrix and the fast Fourier transform to construct the descriptor vector set, accurately extract the image features. The primary registration determination uses the Euclidean distance to screen out possible pre-matching of feature points. The secondary registration determination statistically counts the number of pixel points through the sliding window. The two-stage screening is beneficial to improving the registration accuracy.

[0093] Those skilled in the art should understand that the embodiments of the present invention may provide a method, a system, or a computer program product. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. Among them, the storage medium may be implemented based on any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the functions specified in one process Figure 1 one process or multiple processes and / or boxes Figure 1 the functions specified in one box or multiple boxes.

[0094] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. A method for processing UAV remote sensing mapping images, characterized in that: include: Step S1: collecting image data of the target area to obtain a mapping image set to be processed; Step S2: grayscale mapping is performed on the mapping image set to be processed to obtain an initial grayscale image set, and the initial grayscale image set is denoised by median filtering with a window weight factor. Step S3: calculating and processing the denoised grayscale image by inverse grayscale transformation and color channel restoration algorithm to obtain a color enhanced mapping image; Step S4: Deblurring the color enhanced surveying image, calculating the grayscale main value area representative, the fixed segmentation threshold and the ambient light interference amount based on the denoised grayscale image and the color enhanced surveying image, and obtaining the deblurred surveying image through dynamic segmentation threshold calculation; Step S5: performing registration and stitching processing based on the deblurred remote sensing mapping image to obtain a panoramic registration image; The color enhanced mapping image is deblurred to obtain a deblurred mapping image, and the processing logic includes: Compare the R channel, G channel and B channel values ​​of each pixel in the color enhanced mapping image and take the minimum value to obtain the dark channel value, traverse all the pixels in the color enhanced mapping image to obtain the dark channel mapping image, sort the pixel values ​​in the dark channel mapping image from high to low, select the largest U% pixel value in the dark channel mapping image based on the preset ambient light ratio coefficient U% and sum them to obtain the ambient light interference amount; A grayscale histogram is calculated based on the denoised grayscale image, and a grayscale probability distribution is obtained by linearly mapping the grayscale histogram through the Min-Max normalization algorithm. The mutual correlation coefficient between the grayscale probability distribution and the standard normal distribution function is calculated, and the grayscale probability distribution is threshold-screened through a preset mutual correlation parameter. The grayscale values ​​in the grayscale probability distribution that are greater than or equal to the preset mutual correlation parameter are selected as the grayscale main value area, and the grayscale main value area is median-processed to obtain the grayscale main value area representative; The grayscale histogram is calculated by Otsu's algorithm to obtain a fixed segmentation threshold, and the dynamic segmentation threshold is calculated based on the grayscale main value area representative, the fixed segmentation threshold and the ambient light interference amount; Its calculation expression is: ； Among them, DST represents the dynamic segmentation threshold, max(·) represents the maximum value function, ALI represents the ambient light interference, PGR represents the grayscale main value region representative, and BST represents the fixed segmentation threshold.

2. The method for processing UAV remote sensing images as claimed in claim 1, characterized in that: The UAV is controlled to fly based on the planned inspection route, and the remote sensing optical sensor on the UAV platform collects image data of the target mapping area to obtain the mapping image set to be processed.

3. The method for processing UAV remote sensing imaging images as claimed in claim 1, characterized in that: Grayscale mapping is performed on each image in the mapped image set to be processed to obtain an initial grayscale image set, and the initial grayscale image set is partitioned by a sliding window of preset size and step length to obtain window sub-images. The window grayscale median corresponding to each window sub-image is obtained by sorting calculation, and the window weight factor is calculated based on the window grayscale median corresponding to each window sub-image.

4. The method for processing UAV remote sensing imaging images as claimed in claim 3, characterized in that: The processing logic for calculating the window weight factor includes: The grayscale average value of all pixels in each window sub-image is obtained by an averaging calculation. The grayscale average value corresponding to each window sub-image and the window grayscale median are respectively subtracted and the absolute value is taken to obtain the pixel discrete data. The discrete data corresponding to each window sub-image is averaged twice to obtain the window discrete data. The window weight factor is obtained based on the pixel discrete data and the window discrete data. The corresponding sliding window is weighted based on the window weight factor to obtain the denoised grayscale image. The calculation expression is: ； ； ； Among them, x represents the horizontal coordinate of the pixel point, y represents the vertical coordinate of the pixel point, and G(x,y) represents the grayscale value of the pixel point at the coordinate (x,y). represents the grayscale average value of the window sub-image, DAV(x,y) represents the discrete data of the pixel points, AAV(x,y) represents the discrete data of the window, and N represents the total number of pixels of the window sub-image. represents the window weight factor, c represents the window weight factor constant, and max{·} represents the maximum value function.

5. The method for processing UAV remote sensing imaging images as claimed in claim 1, characterized in that: The denoised grayscale image is inversely transformed to obtain a denoised original color mapping image, and the denoised original color mapping image is computationally processed using an improved color channel restoration algorithm to obtain a color enhanced mapping image.

6. The method for processing UAV remote sensing imaging images as claimed in claim 5, characterized in that: The processing logic of the improved color channel restoration algorithm includes: Based on the preset first color depth value, the denoised original color mapping image is subjected to color quantization reduction processing to obtain a low-bit color depth image, the low-bit color depth image is subjected to color enhancement calculation to obtain a transition-enhanced image, and based on the preset second color depth value, the transition-enhanced image is subjected to color quantization enhancement processing to obtain a color-enhanced mapping image, and the calculation expression thereof is: ； ； Among them, x represents the horizontal coordinate of the pixel, y represents the vertical coordinate of the pixel, PT(x,y) represents the transition enhanced image, PI(x,y) represents the low-bit color depth image, e represents the natural constant, MF(x,y) represents the color enhancement function, Represents a mapping constant, represents the gain constant, BC(x,y) represents the color band in the low-bit color depth image, and M represents the total number of color bands in the low-bit color depth image.

7. The method for processing UAV remote sensing imaging images as claimed in claim 1, characterized in that: The denoised grayscale image is segmented based on a dynamic segmentation threshold to obtain a first blurred area image and a first non-blurred area image, and the dark channel mapping image is subjected to mask segmentation processing based on the pixel coordinates of the first blurred area image and the first non-blurred area image to obtain a second blurred area image and a second non-blurred area image; A deblurring dynamic coefficient is calculated based on the first fuzzy area image and the dynamic segmentation threshold, and the deblurring dynamic coefficient and the second fuzzy area image are multiplied to obtain a fuzzy area enhanced image; Based on the blurred area enhanced image and the second non-blurred area image, an image stitching process is performed to obtain a deblurred dark channel image, and based on the atmospheric light scattering model and the deblurred dark channel image, a restoration process is performed to obtain a deblurred remote sensing mapping image; The calculation expression of the defuzzification dynamic coefficient is: ； Among them, x represents the horizontal coordinate of the pixel point, y represents the vertical coordinate of the pixel point, and DEV(x,y) represents the deblurring dynamic coefficient. represents the limiting constant, FGR(x,y) represents the grayscale value of the pixel in the first fuzzy area image, and DST represents the dynamic segmentation threshold.

8. The method for processing UAV remote sensing imaging images as claimed in claim 1, characterized in that: Based on the deblurred remote sensing mapping image, a reference image and an image to be registered are selected, and the reference image and the image to be registered are registered and spliced ​​to obtain a panoramic registered image.

9. The method for processing UAV remote sensing imaging images as claimed in claim 1, characterized in that: The processing logic of performing registration processing on the image to be registered to obtain the panoramic registration image includes: The first corner point feature vector group and the second corner point feature vector group are obtained by calculating the reference image and the image to be registered respectively through the Harris matrix; Performing FFT calculation on the reference image and the image to be registered by fast Fourier transform to obtain a first phase angle data set and a second phase angle data set, extracting frequency domain phase data from the first phase angle data set and connecting them in series with the corresponding vectors in the first corner point feature vector group to obtain a first descriptor vector set, extracting frequency domain phase data from the second phase angle data set and connecting them in series with the corresponding vectors in the second corner point feature vector group to obtain a second descriptor vector set; Perform a registration determination based on the first descriptor vector set and the second descriptor vector set, calculate the Euclidean distance value between the elements in the first descriptor vector set and the second descriptor vector set, and determine that the feature points are pre-paired when the Euclidean distance is greater than or equal to a preset first registration threshold; Based on the pre-pairing of feature points, a secondary registration judgment is performed, and the number of pixel points that form the pre-pairing of feature points in the image to be registered is counted through a registration sliding window of size Q*Q. When the number of pixel points that form the pre-pairing of feature points in the image to be registered is greater than or equal to a preset second registration threshold, the feature points are retained to obtain the final pairing; Based on the final alignment of the feature points of the reference image and the image to be registered, the registration calculation is performed to the panoramic registration image through the affine transformation model.

Images