Orthorectified Image Mosaicking Method and System Applicable to Remote Sensing Images

Abstract

This application relates to the field of image processing technology, specifically to an orthophoto mosaicking method and system applicable to remote sensing images, which specifically includes: analyzing the similarity of luminance values between data points in multi-spectral remote sensing images, obtaining the ratio relationship between different regions, so as to achieve precise division of different regions in the image, and further obtaining the difference values between these regions. Furthermore, during the image mosaicking process, targeted correction is performed on low-luminance regions according to the obtained difference values to improve the overall effect of image fusion mosaicking; in the orthophoto generated by processing mountain remote sensing images, the luminance difference between the sunny region and the shady region can be further used to determine the adjustment amplitude required for the shady region during HSV fusion mosaicking, so as to be able to adaptively adjust the luminance of the shady region, combine with the high-resolution panchromatic orthophoto to obtain a high-resolution multi-spectral orthophoto with adjusted luminance, and then perform orthophoto mosaicking to improve the fusion effect of the mosaicked image.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and specifically to an orthoimage mosaicking method and system applicable to remote sensing images. Background Art

[0002] Remote sensing images provide great convenience in terrain interpretation. By analyzing the spectral differences between different data points in multi-spectral remote sensing images, different terrain regions can be effectively interpreted and divided. However, in actual use, due to the large terrain undulation in mountainous areas, the brightness difference between the sunny and shady areas of mountains and peaks is significant, which may lead to unnatural phenomena during the image fusion process, thus affecting the interpretation effect of the images.

[0003] When the existing technology uses the HSV fusion method to perform fusion mosaicking analysis on multi-spectral remote sensing orthoimages in mountainous areas, there are defects that due to the large brightness difference between the shady and sunny areas of the mountains, these areas show unnatural effects in the fused images, and thus the terrain interpretation is inaccurate. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide an orthoimage mosaicking method and system applicable to remote sensing images, and the specific technical solutions adopted are as follows:

[0005] In the first aspect, an embodiment of this application provides an orthoimage mosaicking method applicable to remote sensing images, and this method includes the following steps:

[0006] Collect each multi-spectral remote sensing image and the corresponding high-resolution panchromatic image, and perform geometric correction to obtain each multi-spectral orthoimage and high-resolution panchromatic orthoimage;

[0007] For each multi-spectral orthoimage, obtain the brightness image of the multi-spectral orthoimage in the HSV color space, perform region segmentation to obtain each region in the brightness image; obtain the surrounding brightness difference of each region in the brightness image based on the brightness difference between each region and its adjacent regions; distinguish each normal region and each abnormal region in the brightness image through the surrounding brightness difference of each region;

[0008] Obtain each band image of the multi-spectral orthoimage; obtain the state parameter of each region in each band image based on the spectral wavelength and the corresponding brightness value of the pixel points in each region of each band image; construct the wavelength influence coefficient of each region in each band image based on the state parameter difference between each region and its adjacent region in each band image, in combination with the surrounding brightness difference of each region.

[0009] Based on the wavelength influence coefficients of each region in any two adjacent band images, combined with the wavelength differences between the adjacent bands, construct the difference coefficients between any two adjacent band images for each region; Denote the mean value of all the difference coefficients of all normal regions as the normal difference value, and combined with the difference coefficients of each abnormal region, construct the adjustment amplitude coefficients that need to be adjusted for each abnormal region.

[0010] Based on the amplitude coefficients of each abnormal region and the pixel values of pixel points in each band image, construct the corrected pixel values of each channel for each pixel point in each abnormal region, and combined with the high-resolution panchromatic orthophoto, obtain the high-resolution multispectral orthophoto with brightness correction; Mosaic all the high-resolution multispectral orthophotos.

[0011] In one embodiment, the expression for the surrounding brightness difference of each region in the brightness image is:

[0012] , where is the surrounding brightness difference of the th region, is the number of adjacent regions of the th region, is the brightness mean value of all pixel points in the th region, is the brightness mean value of all pixel points in the th adjacent region of the th region, is the normalization function.

[0013] In one embodiment, the state parameter of each region in each band image is: Calculate the ratio between the spectral wavelength and the corresponding brightness value of each pixel point in each region, and denote it as the first ratio; Take the sum value of the first ratios of all pixel points in each region as the state parameter of each region.

[0014] In one embodiment, the process of obtaining the wavelength influence coefficient of each region in each band image is:

[0015] In each band image, calculate the mean value of the state parameters of all adjacent regions of any region; Calculate the difference between the mean value and the state parameter of the any region, and denote it as the first difference; Take the product of the first difference and the surrounding brightness difference of the any region as the wavelength influence coefficient of the any region.

[0016] In one embodiment, the process of obtaining the difference coefficient between any two adjacent band images for each region is:

[0017] The ratio of the wavelength influence coefficient of each region in an image of any wavelength band to the wavelength influence coefficient of the image in the subsequent wavelength band of the any wavelength band image is denoted as the second ratio; calculate the difference amount between the central wavelength of the any wavelength band and the central wavelength of the corresponding subsequent wavelength band; use the normalized value of the product of the second ratio and the difference amount as the difference coefficient of each region between the image of the any wavelength band and the corresponding subsequent wavelength band image.

[0018] In one embodiment, the expression for the adjustment amplitude coefficient that needs to be adjusted for each abnormal region is:

[0019] , where is the adjustment amplitude coefficient that needs to be adjusted for the th abnormal region, is the mean value of all the difference coefficients of the th abnormal region, and is the normal difference value.

[0020] In one embodiment, the process of obtaining the corrected pixel values of each channel for each pixel point in each abnormal region is as follows:

[0021] Denote the corrected channel pixel value of the th pixel point in the th abnormal region as , and the expression of is: , where is the normalized value of the original pixel value of the th pixel point in the th abnormal region in the R-band image, and is the adjustment amplitude coefficient that needs to be adjusted for the th abnormal region;

[0022] Based on the pixel values of each pixel point in the G and B band images and the amplitude coefficients of each abnormal region, respectively, and using the same acquisition method as the corrected R channel pixel value, obtain the corrected G and B channel pixel values of each pixel point in each abnormal region.

[0023] In one embodiment, the process of obtaining the high-resolution multispectral orthophoto image after brightness correction is as follows:

[0024] Based on the corrected R, G, and B channel pixel values of each pixel point in each abnormal region, perform HSV color space conversion to obtain the corrected brightness values of each pixel point in each abnormal region; use the high-resolution panchromatic orthophoto image to replace the V channel image in the HSV image;

[0025] Based on the corrected brightness value and the replaced V-channel image, a modified V-channel image is obtained, and then a modified HSV image is obtained; the modified HSV image is transformed back into the RGB color space to obtain a high-resolution multispectral orthophoto with corrected brightness.

[0026] In one embodiment, the process of obtaining the modified V-channel image is as follows: for each pixel point in each abnormal region of the replaced V-channel image, the original brightness value of each pixel point is replaced with the corrected brightness value of each pixel point to obtain the modified V-channel image.

[0027] In a second aspect, an orthophoto mosaicing system applicable to remote sensing images provided by an embodiment of the present application includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0028] The embodiments of the present application have at least the following beneficial effects:

[0029] By analyzing the similarity of brightness values between data points in a multispectral remote sensing image, the present application obtains the ratio relationship between different regions, thereby achieving precise division of different regions in the image, and further obtaining the difference values between these regions. Then, in the process of image mosaicing, according to the obtained difference values, targeted correction is performed on low-brightness regions to improve the overall effect of image fusion mosaicing; by processing the brightness difference between the sunny region and the shady region in the orthophoto generated from mountain remote sensing images, the adjustment amplitude required for the shady region can be further determined when performing HSV fusion mosaicing, so as to be able to adaptively adjust the brightness of the shady region, combine with the high-resolution panchromatic orthophoto to obtain a high-resolution multispectral orthophoto with adjusted brightness, and then perform orthophoto mosaicing to improve the fusion effect of the mosaiced image. Description of the Drawings

[0030] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0031] Figure 1 It is a flowchart of the steps of an orthophoto mosaicing method applicable to remote sensing images provided by an embodiment of the present application;

[0032] Figure 2 It is a schematic diagram of the process of obtaining a high-resolution multispectral orthophoto with corrected brightness. Detailed Embodiments

[0033] In order to further elaborate on the technical means and effects adopted by this application to achieve the intended invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, detail the orthoimage mosaicking method and system applicable to remote sensing images proposed according to this application, including its specific implementation manners, structures, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs.

[0035] The following will specifically describe the specific solutions of the orthoimage mosaicking method and system applicable to remote sensing images provided by this application in conjunction with the accompanying drawings.

[0036] Please refer to Figure 1 , which shows a flowchart of the steps of an orthoimage mosaicking method applicable to remote sensing images provided by an embodiment of this application. The method includes the following steps:

[0037] Step S1, collect each multispectral remote sensing image and the corresponding high-resolution panchromatic image, and perform geometric correction to obtain each multispectral orthoimage and high-resolution panchromatic orthoimage.

[0038] Collect multispectral remote sensing images of mountainous areas through an unmanned aerial vehicle equipped with a multispectral imager. At the same time, collect high-resolution panchromatic images of mountainous areas through satellite sensors. Preferably, in the embodiments of this application, two multispectral remote sensing images with overlapping areas are used as the multispectral remote sensing images to be mosaicked. As other embodiments of this application, implementers can set the multispectral remote sensing images to be mosaicked according to actual situations.

[0039] Due to the influence of the shooting angle and terrain undulation, there are inclination phenomena in mountains, peaks, etc. in the collected images. Therefore, the collected multispectral remote sensing images and high-resolution panchromatic images are geometrically corrected through the RPC (Rational Polynomial Coefficients) model to obtain multispectral orthoimages and high-resolution panchromatic orthoimages. Among them, the RPC model is a well-known technology, and the specific process will not be elaborated here.

[0040] It should be noted that for the acquisition of orthoimages, this application only provides a remote sensing image correction method. There are many existing remote sensing image correction methods, and implementers can also use other remote sensing image correction models to obtain multispectral orthoimages and high-resolution panchromatic orthoimages. This application does not make specific restrictions.

[0041] Step S2: For each multispectral orthophoto, obtain the brightness image of the multispectral orthophoto in the HSV color space, perform region segmentation to obtain each region in the brightness image; obtain the surrounding brightness difference of each region in the brightness image based on the brightness difference between each region and its adjacent regions; distinguish each normal region and each abnormal region in the brightness image through the surrounding brightness difference of each region.

[0042] Taking any multispectral orthophoto as an example, the following analysis is carried out:

[0043] (1) Obtain the brightness image of the multispectral orthophoto:

[0044] Convert the multispectral orthophoto to the HSV color space, and use the V-channel image in the HSV color space as the brightness image of the multispectral orthophoto. Among them, the conversion of the HSV color space is a well-known technology, and the specific process will not be elaborated here.

[0045] It should be noted that for the acquisition of the brightness image of the orthophoto, this application only provides a method for obtaining the brightness image. There are many existing methods for obtaining the brightness image, and the implementer can also use other brightness conversion methods to obtain the brightness image of the orthophoto. This application does not make specific restrictions.

[0046] (2) Obtain each region in the brightness image of the multispectral orthophoto:

[0047] Perform contour detection on the brightness image of the multispectral orthophoto, take the number of regions detected in the brightness image as the number of initial seed points of the region growing algorithm, and further use the brightness image as the input of the region growing algorithm. The output is each region in the brightness image, realizing the region segmentation of the brightness image.

[0048] It should be noted that for the region segmentation of the brightness image, this application only provides a region segmentation method. There are many existing region segmentation methods, and the implementer can also use other region segmentation algorithms to perform region segmentation on the brightness image. This application does not make specific restrictions.

[0049] (3) Obtain the surrounding brightness difference of each region in the brightness image of the multispectral orthophoto:

[0050] Due to the obstruction of mountains, the shady areas and valleys in mountainous areas receive less sunlight, resulting in these terrain areas appearing darker. In the images obtained through aerial data, the brightness of these areas is significantly lower than that of adjacent areas. Therefore, for the brightness image of the multispectral orthophoto, based on the difference between the brightness of each region in the brightness image and the brightness of each of its adjacent regions, construct the surrounding brightness difference of each region in the brightness image. The expression is:

[0051] , where, is the surrounding brightness difference of the th region, is the number of adjacent regions of the th region, is the average brightness of all pixel points in the th region, is the average brightness of all pixel points in the th region and the th adjacent region of the is the normalization function.

[0052] The greater the brightness difference between each region and its adjacent regions, and at the same time, the lower the brightness of each region, the more likely it is that the region is a shaded area, a valley, or other areas with low sunlight intensity, and the more necessary it is to adjust the brightness of the region.

[0053] (4) Set a brightness difference threshold. Preferably, in the embodiments of the present application, the brightness difference threshold is set to 0.6. As other embodiments of the present application, the implementer can set the brightness difference threshold according to the actual situation. If any region in the brightness image is greater than or equal to the brightness difference threshold, then the any region is regarded as an abnormal region, and the abnormal region and all its adjacent regions are combined to obtain an abnormal region group, where the abnormal region is recorded as the central abnormal region of its abnormal region group. Similarly, if any region in the brightness image is less than the brightness difference threshold, then the any region is regarded as a normal region, and the normal region and all its adjacent regions are combined to obtain a normal region group, where the normal region is recorded as the central normal region of its normal region group.

[0054] Obtain each abnormal region and each normal region of the multispectral orthophoto image through the above method.

[0055] Step S3, obtain the images of each band of the multispectral orthophoto image; obtain the state parameters of each region in each band image based on the spectral wavelengths and corresponding brightness values of the pixel points in each region of each band image; construct the wavelength influence coefficient of each region in each band image based on the state parameter differences between each region and its adjacent regions in each band image, in combination with the surrounding brightness differences of each region.

[0056] (1) For the multispectral orthophoto image, obtain the image of each band of the multispectral orthophoto image, and each band of the multispectral orthophoto image includes blue (B), green (G), red (R), red edge (RE), and near-infrared (NIR) bands.

[0057] (2)When conducting multispectral imaging in mountainous areas, due to the changes in the incident angle of sunlight and light intensity, the shaded side of the mountains receives less light, resulting in a decrease in the spectral reflectance of the surface of the ground objects in this area. As a result, the shaded area appears darker in the multispectral remote sensing image, and this kind of phenomenon will occur in the spectral images of different bands. Therefore, in order to obtain the influence degree of the shadow area on the spectral information of different bands, in any band image of the multispectral orthophoto image, for any abnormal area group, based on the difference between the spectral wavelength and brightness of each area in the abnormal area group, the state parameter of each area in the abnormal area group is constructed, and the expression is:

[0058] ;

[0059] In the formula, is the state parameter of the th area in the abnormal area group, is the number of pixel points in the th area in the abnormal area group, is the spectral wavelength of the th pixel point in the th area in the abnormal area group, is the brightness value corresponding to the th pixel point in the th area in the abnormal area group in the brightness image. Among them, is the first ratio, indicating the relationship between brightness and overall parameters.

[0060] Furthermore, in order to analyze the actual influence of the wavelength on the terrain change, for any abnormal area group in the said any band image, based on the state parameters of each area in the abnormal area group, combined with the surrounding brightness difference of the central abnormal area of the abnormal area group, the wavelength influence coefficient of the central abnormal area of the abnormal area group is constructed, and the expression is:

[0061] ;

[0062] In the formula, is the wavelength influence coefficient of the central abnormal area of the abnormal area group, is the surrounding brightness difference of the central abnormal area of the abnormal area group, is the number of adjacent areas of the central abnormal area of the abnormal area group, is the state parameter of the th adjacent area of the central abnormal area of the abnormal area group, is the state parameter of the central abnormal area of the abnormal area group, is the normalization function. Among them, is the first difference.

[0063] The first difference is the difference between the state parameters of the abnormal region and those of its adjacent region, indicating the degree of influence of the shadow on the spectra of different wavelengths.

[0064] Through the above method, the wavelength influence coefficients of each abnormal region in any band image of the multispectral remote sensing image can be obtained.

[0065] Step S4: Based on the wavelength influence coefficients of each region in the images of any two adjacent bands, combined with the wavelength difference between the adjacent bands, construct the difference coefficient between the images of any two adjacent bands for each region; Denote the mean value of all the difference coefficients of all normal regions as the normal difference value, and combined with the difference coefficients of each abnormal region, construct the adjustment amplitude coefficient that needs to be performed for each abnormal region.

[0066] (1) The degrees of influence on different wavelengths may vary. It is necessary to analyze the influence on the same region under different wavelengths, and then obtain the degree of influence of different wavelengths on the spectral image. When acquiring the multispectral remote sensing image of the undulating mountain area, by collecting spectral images of multiple bands for the same terrain region and synthesizing the spectral images of multiple bands. During the synthesis process, it is necessary to analyze the smoothness of the fusion. If the spectral images of different bands in the synthesized multispectral remote sensing image can be smoothly fused and the pixels are clear, it meets the fusion standard; otherwise, if the resolution of the fused image is low and blurred, adjustment is required. Therefore, when fusing the spectral images of multiple bands, it is necessary to compare the spectral splicing smoothness between the spectral images of different bands.

[0067] For the images of any two adjacent bands of the multispectral remote sensing image, based on the difference between the wavelength influence coefficients of each abnormal region in the images of these two adjacent bands, combined with the wavelength difference between these two adjacent bands, construct the difference coefficient between the images of these two adjacent bands for each abnormal region. The expression is:

[0068] , where is the difference coefficient between the th abnormal region of the multispectral remote sensing image between the th and the th band images, , are respectively the th abnormal region in the th and the th band images, is the difference amount between the central wavelengths of the th and the th bands, is the sigmoid function. Among them, is the second ratio.

[0069] The larger it is, the greater the degree of influence of adjacent bands by shadows, and vice versa; The larger it is, the greater the actual difference between the two spectra.

[0070] (2) When synthesizing multi-spectral remote sensing images from spectral images of multiple bands, relying solely on fixed thresholds or single fixed values cannot accurately reflect the specific degree of influence of different bands by shadows. Therefore, first, based on the spectral wavelengths and corresponding brightness values of all pixel points in each normal region group in all band images, as well as the brightness difference around the central normal region, using the same acquisition method as the difference coefficient between any two adjacent band images of each abnormal region, the difference coefficient between any two adjacent band images of each normal region is obtained.

[0071] Then, calculate the mean value of the difference coefficients of all normal regions between all adjacent band images, denoted as the normal difference value.

[0072] After that, based on the difference coefficient between any two adjacent band images of each abnormal region, combined with the normal difference value, construct the adjustment amplitude coefficient that needs to be adjusted for each abnormal region, and the expression is:

[0073] , where in the formula, is the adjustment amplitude coefficient that needs to be adjusted for the th abnormal region, is the mean value of all the difference coefficients of the th abnormal region, is the normal difference value.

[0074] Step S5, based on the amplitude coefficient of each abnormal region and the pixel values of the pixel points in each band image, construct the corrected pixel values of each channel of each pixel point in each abnormal region, and combined with the high-resolution panchromatic orthoimage, obtain the high-resolution multi-spectral orthoimage with corrected brightness; mosaic all high-resolution multi-spectral orthoimages.

[0075] (1) Usually, the R, G, and B band images of multi-spectral remote sensing images are superimposed to make the visualized color closer to the RGB image of the actual scene. Therefore, for the R band image of the multi-spectral image, based on the pixel values of each pixel point in the R band image and the adjustment amplitude coefficient that needs to be adjusted for each abnormal region, construct the corrected pixel values of each pixel point in each abnormal region of the R band image, and the expression is:

[0076] , where in the formula, is the th pixel point in the The corrected R-channel pixel value of a pixel point, is the normalized value of the original pixel value of the pixel point in the th abnormal area in the R-band image, is the adjustment amplitude coefficient that needs to be adjusted for the th abnormal area.

[0077] Based on the pixel values of each pixel point in the G and B band images and the adjustment amplitude coefficients that need to be adjusted for each abnormal area, using the same acquisition method as the corrected R-channel pixel value, the corrected G and B channel pixel values of each pixel point in each abnormal area are obtained.

[0078] (2) Perform HSV color space conversion based on the corrected R, G, and B channel pixel values of each pixel point in each abnormal area to obtain the corrected brightness values of each pixel point in each abnormal area. Among them, obtaining the V value of each pixel point in the HSV space based on the RGB values of each pixel point is a well-known technology, and the specific process will not be elaborated.

[0079] (3) For the HSV image of the multispectral remote sensing image, first use the high-resolution panchromatic orthophoto image to replace the V-channel image in the HSV image, and then for each pixel point in each abnormal area of the replaced V-channel image, use the corrected brightness value of each pixel point to replace the original brightness value of each pixel point to obtain the modified V-channel image, and then obtain the modified HSV image.

[0080] Furthermore, transform the modified HSV image back to the RGB color space, and the obtained RGB image is the high-resolution multispectral orthophoto image after brightness correction. At this time, the obtained RGB image contains both the spectral information of the original multispectral remote sensing image and the spatial resolution of the high-resolution panchromatic image.

[0081] (4) Perform the above processing on each multispectral remote sensing image to be mosaicked to obtain each high-resolution multispectral orthophoto image, and mosaic all the high-resolution multispectral orthophoto images to form a larger-range mosaic image to improve the correctness of the analysis result and complete the mosaic of multiple remote sensing images. Among them, mosaicking the high-resolution multispectral orthophoto images is a well-known technology, and the specific process will not be elaborated.

[0082] The schematic diagram of the process of obtaining the high-resolution multispectral orthophoto image after brightness correction is as shown in Figure 2 Figure.

[0083] Based on the same inventive concept as the above method, an orthophoto mosaic system applicable to remote sensing images is further provided in an embodiment of the present application, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above methods for orthophoto mosaic applicable to remote sensing images are implemented.

[0084] In summary, an embodiment of the present application provides a method for orthophoto mosaic applicable to remote sensing images. By analyzing the similarity of luminance values between data points in a multispectral remote sensing image, the ratio relationship between different regions is obtained, so as to achieve precise division of different regions in the image, and further obtain the difference values between these regions. Furthermore, during the image mosaic process, the low-luminance regions are targeted for correction according to the obtained difference values to improve the overall effect of image fusion and mosaic. In the orthophoto generated by processing mountain remote sensing images, the luminance difference between the sunny region and the shaded region can be further determined to obtain the adjustment amplitude required for the shaded region during HSV fusion and mosaic, so that the luminance of the shaded region can be adaptively adjusted. Combining with the high-resolution panchromatic orthophoto, a high-resolution multispectral orthophoto with adjusted luminance is obtained, and then orthophoto mosaic is performed to improve the fusion effect of the mosaicked image.

[0085] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of the present application have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0086] The various embodiments in the present application are all described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

[0087] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present application shall be included in the protection scope of the present application.

Claims

1. An orthophoto mosaic method applicable to remote sensing images, characterized in that: The method comprises the following steps: Collect various multispectral remote sensing images and corresponding high-resolution panchromatic images, and perform geometric correction to obtain various multispectral orthophotos and high-resolution panchromatic orthophotos; For each multispectral orthophoto, a brightness image of the multispectral orthophoto in the HSV color space is obtained, and regional segmentation is performed to obtain each region in the brightness image; based on the brightness difference between each region and the adjacent region, the surrounding brightness difference of each region in the brightness image is obtained; and the normal regions and abnormal regions in the brightness image are distinguished by the surrounding brightness difference of each region; Obtain each band image of the multispectral orthophoto; obtain the state parameters of each area in each band image based on the spectral wavelength and corresponding brightness value of the pixel points in each area in each band image; construct the wavelength influence coefficient of each area in each band image based on the difference in state parameters between each area in each band image and its adjacent areas, combined with the surrounding brightness difference of each area; Based on the wavelength influence coefficient of each region in any two adjacent band images, combined with the wavelength difference of the adjacent bands, the difference coefficient of each region between any two adjacent band images is constructed; the average of all the difference coefficients of all normal regions is recorded as the normal difference value, and combined with the difference coefficients of each abnormal region, the amplitude coefficient that needs to be adjusted for each abnormal region is constructed; Based on the amplitude coefficient of each abnormal area and the pixel value of each pixel in each band image, the corrected pixel value of each channel of each pixel in each abnormal area is constructed, and combined with the high-resolution panchromatic orthophoto, the high-resolution multispectral orthophoto with brightness correction is obtained; all high-resolution multispectral orthophotos are mosaicked.

2. The orthophoto mosaic method applicable to remote sensing images as claimed in claim 1, characterized in that: The expression of the surrounding brightness difference of each area in the brightness image is: , where For the The surrounding brightness difference of each area, For the The number of adjacent regions of a region, For the The average brightness of all pixels in the region, For the Region The average brightness of all pixels in the adjacent area, is the normalization function.

3. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The state parameter of each area in each band image is as follows: the ratio between the spectral wavelength of each pixel point in each area and the corresponding brightness value is calculated, recorded as the first ratio; and the sum of the first ratios of all pixels in each area is taken as the state parameter of each area.

4. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The process of obtaining the wavelength influence coefficient of each area in each band image is as follows: In each band image, the mean of the state parameters of all adjacent areas of any area is calculated; the difference between the mean and the state parameter of any area is calculated and recorded as the first difference; the product of the first difference and the surrounding brightness difference of any area is taken as the wavelength influence coefficient of any area.

5. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The process of obtaining the difference coefficient between any two adjacent band images of each region is as follows: The ratio of the wavelength influence coefficient of each region in any band image to the wavelength influence coefficient in the subsequent band image of the any band image is recorded as the second ratio; the difference between the central wavelengths of the any band and the corresponding subsequent band is calculated; and the normalized value of the product of the second ratio and the difference is used as the difference coefficient of each region between the any band image and the corresponding subsequent band image.

6. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The expression of the amplitude coefficient that needs to be adjusted in each abnormal area is: , where For the The amplitude coefficient of the abnormal area that needs to be adjusted, For the The mean of all the coefficients of variation in the abnormal regions, is the normal difference value.

7. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The process of obtaining the corrected pixel values ​​of each channel of each pixel point in each abnormal area is as follows: The first In the abnormal area The corrected channel pixel value of the pixel is recorded as , The expression is: , where The R band image In the abnormal area The normalized value of the original pixel value of pixels, For the The amplitude coefficient of each abnormal area that needs to be adjusted; Based on the pixel value of each pixel point in the G and B band images and the amplitude coefficient of each abnormal area, the corrected G and B channel pixel values ​​of each pixel point in each abnormal area are obtained in the same manner as the corrected R channel pixel value.

8. The orthophoto mosaic method applicable to remote sensing images according to claim 1, characterized in that: The process of obtaining the brightness-corrected high-resolution multispectral orthophoto is as follows: Based on the corrected R, G, and B channel pixel values ​​of each pixel in each abnormal area, HSV color space conversion is performed to obtain the corrected brightness value of each pixel in each abnormal area; Replace the V channel image in the HSV image with a high-resolution panchromatic orthophoto; A modified V channel image is obtained based on the corrected brightness value and the replaced V channel image, and then a modified HSV image is obtained; the modified HSV image is transformed back into the RGB color space to obtain a high-resolution multispectral orthophoto with corrected brightness.

9. The orthophoto mosaic method applicable to remote sensing images as claimed in claim 8, characterized in that: The modified V channel image acquisition process is: for each pixel point in each abnormal area in the replaced V channel image, the original brightness value of each pixel point is replaced by the corrected brightness value of each pixel point to obtain the modified V channel image.

10. An orthophoto mosaic system for remote sensing images, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images