Color uniformizing method and system based on original remote sensing image and electronic equipment

Abstract

The invention provides a color homogenizing method and system based on an original remote sensing image and electronic equipment, and the method comprises the steps: carrying out the geometric correction of the original remote sensing image, and carrying out the resampling; carrying out primary color homogenization on the resampled original remote sensing image, and then carrying out mosaic processing to generate a template image; blocking the original remote sensing image according to a preset size and an overlapping degree, and recording a starting point and width and height information of each blocked image; calculating latitude and longitude corresponding areas of the block images in the template image; calculating a mean value and a variance of latitude and longitude corresponding areas in the block image and the template image, interpolating the mean value and the variance of each pixel according to the calculated mean value and the calculated variance, and obtaining a mean value image and a variance image with the same size as the original image and the template image; and performing fine color homogenization on the original remote sensing image pixel by pixel in combination with the interpolated mean value image, the interpolated variance image, the reference mean value image and the reference variance image. According to the invention, color uniformity of the original remote sensing image is realized by means of a computer technology.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and more specifically, to a color homogenization method, system and electronic device based on original remote sensing images. Background Art

[0002] Currently, the three-dimensional modeling of satellite images mainly uses original remote sensing images as input. Due to reasons such as differences in time phase, weather, sensors, and terrain undulation, the original images will have uneven illumination and color distribution, which is not conducive to capturing terrain and feature characteristics during the three-dimensional modeling process, resulting in low data quality and poor visual effects of the three-dimensional model. Therefore, color homogenization of original remote sensing images is of great significance for improving the accuracy and efficiency of three-dimensional modeling, improving the model quality, and supporting subsequent analysis applications.

[0003] Existing image color homogenization methods are mainly based on result data such as Digital Orthophoto Map (DOM); these result data have spatial reference information, while original remote sensing images do not have spatial reference information, and it is difficult to process original remote sensing images with such methods. Summary of the Invention

[0004] Based on the technical problems existing in the prior art, the present invention provides a color homogenization method, system and electronic device based on original remote sensing images. By adopting block calculation and using RPC coefficients to convert image pixel coordinates into geographic coordinates, and using the corresponding regions of the divided blocks in the reference template image as the color homogenization background, fine color homogenization processing is performed on the original image to solve the problem that it is difficult to color the original remote sensing image with existing methods.

[0005] According to the first aspect of the present invention, there is provided a color homogenization method based on original remote sensing images, including the following steps: Step1. Geometrically correct the original remote sensing image to obtain a corrected image; Step2. Resample the corrected image according to the target resolution; Step3. Perform preliminary color homogenization on the resampled image, perform mosaicking processing on the color homogenization result, and generate a template image; Step4. Divide the original remote sensing image into blocks according to a preset size and overlap degree, and record the starting points and width and height information of each block image; Step5. Calculate the longitude and latitude corresponding regions of the block image in the template image; Step6. Calculate the mean and variance of the longitude and latitude corresponding regions in the segmented image and the template image. According to the calculated mean and variance, use the bilinear interpolation method to interpolate the mean and variance of each pixel, and obtain a mean image and a variance image with the same size as the original remote sensing image. Perform the same operation on the template image to obtain a reference mean image and a reference variance image with the same size as the original remote sensing image; Step7. Combine the interpolated mean image, variance image, reference mean image, and variance image to perform fine color homogenization on the original remote sensing image pixel by pixel.

[0006] Based on the above technical solution, the present invention can also be improved as follows.

[0007] Optionally, the geometric correction of the original remote sensing image includes: Perform vegetation enhancement on each original remote sensing image, and use rational polynomial coefficients to perform geometric correction on the original remote sensing image to obtain a corrected image, and the corresponding corrected image coordinates are geographic coordinates.

[0008] Optionally, the resampling of the corrected image according to the target resolution includes: Statistically analyze the resolution of the corrected image to obtain the minimum resolution. Multiply the minimum resolution by the coefficient N as the resampling target resolution, and resample the corrected image to make the resolutions of all corrected images consistent.

[0009] Optionally, the coefficient N is obtained by dividing the brightness mean of the minimum resolution image by the brightness variance of the minimum resolution image and then multiplying by a preset value.

[0010] Optionally, the preliminary color homogenization of the resampled image and the mosaic processing of the color homogenization result include: Unify the resolutions of all corrected images, perform preliminary color homogenization on the resampled image, reduce the color difference between all resampled images, and perform mosaic processing on the preliminarily color homogenized image to generate a template image.

[0011] Optionally, the calculation of the longitude and latitude corresponding regions of the segmented image in the template image includes: Use the rational polynomial coefficients of the original remote sensing image data to convert the starting point of each segmented image from pixel coordinates to geographic coordinates, and obtain the corresponding region of the segmented image on the template image according to the converted geographic coordinates.

[0012] Optionally, the calculation of the mean and variance of the longitude and latitude corresponding regions in the segmented image and the template image includes: Statistically analyze the pixel values of the segmented image and the segmented reference image by band respectively. According to the pixel values, calculate the mean and variance of each band to obtain the corresponding variance image and mean image.

[0013] Optionally, the fine color equalization of the original remote sensing image pixel by pixel includes: Taking the template image as the reference target, traversing the pixels of the original remote sensing image by band, converting the original remote sensing image from pixel coordinates to geographic coordinates using rational polynomial coefficients, and obtaining the corresponding position of each pixel coordinate in the original remote sensing image in the template image based on the geographic coordinates.

[0014] According to the second aspect of the present invention, a color equalization system for an original remote sensing image is provided, including: An image loading module for reading and parsing remote sensing data to obtain basic information such as rational polynomial coefficients, image width and height, and image starting point; A geometric correction module for geometrically correcting the image using the obtained rational polynomial coefficients and publicly provided elevation information, so that the coordinates of the original remote sensing image are converted from pixel coordinates to geographic coordinates; A vegetation enhancement module for processing images including the near-infrared band and the red band to enhance the color of the vegetation area; A resampling module for resampling the resolution of the input remote sensing to make the resolution of the input remote sensing data consistent; A color equalization module for performing color equalization on the original remote sensing data using the input reference image or band reference data value and outputting the color equalization result.

[0015] According to the third aspect of the present invention, an electronic device is provided, including a memory and a processor, and the processor is used to implement the steps of a color equalization method for an original remote sensing image when executing a computer program stored in the memory.

[0016] The technical effects and advantages of the present invention: The present invention discloses a color equalization method, system and electronic device for an original remote sensing image, which realizes the color equalization of the original remote sensing image by means of computer technology. Among them, the object of color equalization is the original remote sensing image without spatial reference information, and there are color differences, inconsistent resolutions, and all are in pixel coordinate systems for these remote sensing images. Through vegetation enhancement and rational polynomial coefficient correction of the original remote sensing image, the image becomes a geographic coordinate system. Further, in order to unify the resolution, the remote sensing image is resampled to the target resolution. In order to reduce the color difference of the image and the resolution is lower than that of the original image, the image is preliminarily color-equalized. After resampling, the resolution and color difference of the image are smaller than those of the original image. The initial color equalization uses the set reference variance and mean, which greatly eliminates the color difference of the image. However, at this time, the remote sensing image is in the geographic coordinate system and cannot be used for 3D modeling. Therefore, finally, taking the preliminarily color-equalized image as the color-equalized background, color equalization, block calculation and variance and mean interpolation of the image are performed on the original remote sensing image, which greatly eliminates the color difference between images and ensures natural color transition between adjacent image blocks.

[0017] Other features and advantages of the present invention will be set forth in the following description, and in part will be obvious from the description, or may be learned by practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the structure particularly pointed out in the specification, claims as well as the drawings. Brief Description of the Drawings

[0018] In order to illustrate the technical solutions in the present invention or the prior art more clearly, the following briefly introduces the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings may be obtained based on these drawings.

[0019] Figure 1 It is a schematic flowchart of template image production provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the original remote sensing image provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the initial color homogenization result, i.e., the template image, provided by an embodiment of the present invention; Figure 4 It is a corresponding area map of the block image in the reference template provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of bilinear interpolation provided by an embodiment of the present invention; Figure 6 It is a schematic flowchart of image color homogenization provided by an embodiment of the present invention; Figure 7 It is a diagram of the final color homogenization result provided by an embodiment of the present invention. Detailed Description of the Embodiments

[0020] To make the objectives, technical solutions and advantages of the present invention clearer, the following clearly and completely describes the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0021] It can be understood that, based on the deficiencies in the background art, an embodiment of the present invention proposes a color homogenization method based on the original remote sensing image, and the process reference Figure 1 , and the method includes: Step1, geometrically correct the original remote sensing image to obtain the corrected image; It should be noted that the original remote sensing image used in the embodiment of the present invention is asFigure 2 As shown, there are color differences and inconsistent resolutions in the original remote sensing images, which are visible to the naked eye and the color differences are significant, making them unusable for modeling.

[0022] Therefore, first of all, geometric correction needs to be performed on the original remote sensing images, specifically including: Vegetation enhancement is performed on each original remote sensing image. First, the normalized difference vegetation index NDVI (Normalized Difference Vegetation Index) is calculated using the brightness values of the red light band and the near-infrared light band, and then different strategies are adopted according to the value of NDVI to enhance the color of the vegetation area. The vegetation enhancement calculation formula is as follows:

[0023]

[0024] Among them, is the brightness value of the red light band, is the brightness value of the near-infrared light band, is the enhancement coefficient, which is set to 0.35 in this embodiment, is the brightness value of the green light band, is the enhanced brightness value of the green light band.

[0025] Then, the original remote sensing image is geometrically corrected using the RPC coefficients to obtain the corrected image. At this time, all the coordinates of the corrected image are geographic coordinates.

[0026] It should be noted that the RPC coefficients (Rational Polynomial Coefficients) are polynomial parameters that describe the mapping relationship between image coordinates (row, column) and ground coordinates (longitude, latitude, elevation), and are the core components of the Rational Function Model (RFM). The correction using RPC coefficients is an extension based on the linear polynomial model, and a non-linear relationship is used to establish the correspondence between the pixel coordinates on the image and the longitude and latitude corresponding to the pixels. By combining the pixel coordinates, elevation data, and parameters such as rotation, scaling, and perspective distortion included in the RPC model, the image pixel coordinates are converted into geographic coordinates. The elevation data in this embodiment is from the publicly available DEM elevation data World_geoid_height_EGM96.img.

[0027] Step2, Resample the corrected image according to the target resolution; There is a problem of inconsistent resolution in the original remote sensing images. When directly mosaicking the images, there will be obvious differences in the visual appearance of images with different resolutions. The fineness and richness of details of high-resolution images are usually higher than those of low-resolution images. Direct mosaicking will result in obvious boundaries, affecting the overall visual effect.

[0028] Therefore, it is necessary to resample the corrected images. The resampling includes: Statistical resolution of the corrected images, obtaining the minimum resolution, multiplying the minimum resolution by the coefficient N as the resampling target resolution, and resampling the corrected images to make the resolutions of all corrected images consistent.

[0029] Furthermore, first parse the corrected images to obtain the image resolutions, and take N times the minimum resolution as the resampling target resolution. For the sake of easy description, N times the minimum resolution is represented by the symbol R, and all corrected images are resampled to the resolution R. Among them, N is related to the amount of information of the remote sensing images, and the calculation formula of the N value is as follows:

[0030] Among them, m is a preset value, and the m value in this embodiment is 5. The value represents the average brightness of the image with the minimum resolution. The value is the brightness variance of the image with the minimum resolution.

[0031] Perform preliminary color homogenization on the resampled images to reduce the color difference between all resampled images; the calculation formula is as follows:

[0032] Among them, Represents the processed pixel value; Respectively represent the variance of the resampled image at the (x, y) position corresponding to the template image, the variance of the resampled image at the (x, y) position; Respectively represent the average value of the resampled image at the (x, y) position corresponding to the template image, the average value of the resampled image at the (x, y) position, Respectively represent the brightness coefficient and the variance expansion coefficient.

[0033] When performing preliminary color homogenization on the resampled images, Both take the value of 1. When processing the red light band, Respectively take the values of 54 and 95; when processing the green light band, Respectively take the values of 39 and 103; when processing the blue light band, Respectively take the values of 38 and 95; Step3, perform preliminary color homogenization on the resampled images, perform mosaic processing on the color homogenization results, and generate a template image; The preliminary color homogenization of the resampled image and the mosaicking process of the color homogenization result include: Make all the corrected images reach the same resolution, perform preliminary color homogenization on the resampled images, reduce the color difference between all resampled images, perform a mosaicking process on the preliminarily color-homogenized images to generate a template image. At this time, the color difference of the template image is small and can be basically ignored.

[0034] It should be noted that the mosaicking process is performed on the original remote sensing images with unified resolution to obtain a low-resolution template image; the template image obtained in the embodiment is as Figure 3 shown. Compared with the original image, the color difference of the template image is reduced and the resolution is decreased, which is used as a reference target for fine color homogenization. In this embodiment, the MULTI_BAND method of the third-party open-source library OpenCV is used for mosaicking.

[0035] Step4, divide the original remote sensing image into blocks according to a preset size and overlap degree, and record the starting point and width and height information of each block image; Dividing the original remote sensing image to be processed into blocks according to a preset size and overlap degree is to solve the problems of inconsistent overall tone and obvious local tone difference between adjacent images. For the sake of simple description, the block images are used to represent the images after the original remote sensing data is divided into blocks.

[0036] In this embodiment, the image is divided into NxM sub-regions with a block size of 256x256, and the starting point coordinates, width, height and other information of the sub-regions are recorded. In order to eliminate the color difference in the adjacent areas between blocks, each image block is expanded by 100 pixels outward, and the actual image block size is 356x356.

[0037] Step5, calculate the corresponding longitude and latitude regions of the block image pixels in the template image; Use the RPC coefficients of the original remote sensing image data to convert the starting point of each block image from pixel coordinates to geographic coordinates; since the template image is in a geographic coordinate system, according to the converted geographic coordinates, the corresponding region of the block image on the template image can be obtained. For the sake of simple description, the block template image is used to represent the corresponding region of the block image in the template image. Refer to the schematic Figure 4 , use the RPC coefficients to find the corresponding region for the block image in the template image.

[0038] Step6, calculate the mean and variance of the longitude and latitude corresponding regions in the block image and the template image. According to the calculated mean and variance, use the bilinear interpolation method to interpolate the mean and variance of each pixel, and obtain a mean map and a variance map with the same size as the original image. The same operation is performed on the template image to obtain a reference mean map and a reference variance map with the same size as the original remote sensing image; Calculating the means and variances of the corresponding regions of longitude and latitude in the block image and the template image includes: Statistically analyze the pixel values of the block image and the block reference image by band respectively. Based on these pixel values, calculate the mean and variance of each band to obtain a variance map and a mean map of size NxM. Then, use the bilinear interpolation method to interpolate the mean and variance of each pixel to obtain a mean map and a variance map of the same size as the original image. The block reference image is processed in the same way to obtain a reference mean map and a reference variance map of the same size as the template image. As Figure 5 shown in the bilinear interpolation schematic diagram, use the bilinear interpolation method to obtain a mean map and a variance map of the same size as the original image; similarly, the block template image is also processed in the same way to calculate the mean and variance, and use the bilinear interpolation method to obtain a reference mean map and a reference variance map of the same size as the template image; the interpolation methods for calculating the mean and variance are not limited to but include the bilinear interpolation method.

[0039] In this embodiment, when calculating the variance and mean, pixels with all-band pixel values of 0 are ignored. Each original image is divided into NxM block images according to the block strategy, and one mean and one variance can be calculated for each block image. Accordingly, one mean map and one variance map of size NxM can be obtained for each image. Use the bilinear interpolation method to interpolate the mean map and the variance map to obtain an interpolated mean map and an interpolated variance map of the same size as the original image. The block reference image is also processed using the same principle to obtain an interpolated reference mean map and an interpolated reference variance map of the same size as the template image.

[0040] Step7, perform fine color homogenization on the original remote sensing image pixel by pixel in combination with the interpolated mean map, variance map, reference mean map, and variance map.

[0041] The process of using the template image to perform color homogenization on the original remote sensing image is as Figure 6 shown. Taking the template image as the reference target, traverse the pixels of the original remote sensing image by band, convert its coordinates from pixel coordinates to geographic coordinates using the RPC coefficients, and based on this principle, for each pixel coordinate in the original remote sensing image, find the corresponding position in the template image. The original remote sensing image corresponds one-to-one with the interpolated mean map and the interpolated variance map. The value at the position of the interpolated mean map is , and the value at the position of the interpolated variance map is . The template image corresponds one-to-one with the interpolated reference mean map and the interpolated reference variance map. The value at the position of the interpolated reference mean map is , and the value at the position of the interpolated reference variance map The value of the position is , and the original image is processed according to the calculation formula in Step 3. At this time take the values of 0.75 and 0.6 respectively. Finally, a computer program written in the C++ programming language automatically performs color homogenization on the original remote sensing image. The final color homogenization result is shown in Figure 7 as shown.

[0042] In summary, a color homogenization method for original remote sensing images provided by an embodiment of the present invention uses RPC (Rational Polynomial Coefficients) coefficients to convert image pixel coordinates into geographic coordinates, uses a template image derived from the original remote sensing image as the color homogenization background, and performs fine color homogenization processing on the original remote sensing image to solve the problem that it is difficult to color the original remote sensing image by existing methods.

[0043] According to the second aspect of the present invention, the present invention provides a color homogenization system based on an original remote sensing image, including: An image loading module for reading and parsing remote sensing data in common formats to obtain basic information such as RPC information, image width and height, and image starting point; A geometric correction module for geometrically correcting the image using the obtained RPC coefficients and publicly provided elevation information, so that the coordinates of the original remote sensing image are converted from pixel coordinates to geographic coordinates; A vegetation enhancement module for processing images including both the near-infrared band and the red band to enhance the color of the vegetation area; A resampling module for resampling the resolution of the input remote sensing to make the resolution of the input remote sensing data consistent; A color homogenization module for performing color homogenization on the original remote sensing data using the input reference image or band reference data value and outputting the color homogenization result.

[0044] It can be understood that a color homogenization system based on an original remote sensing image provided by the present invention corresponds to a color homogenization method for an original remote sensing image provided in the foregoing embodiments. The relevant technical features of a color homogenization system based on an original remote sensing image can refer to the relevant technical features of a color homogenization method for an original remote sensing image, which will not be elaborated here.

[0045] According to a third aspect of the present invention, an embodiment of the present invention provides an electronic device, which may include: a processor, a communications interface, a memory, and a communication bus. Among them, the processor, the communications interface, and the memory complete communication with each other through the communication bus. The processor can call the logic instructions in the memory to execute the implementation steps of the method for color homogenization that takes into account both template image production and the original remote sensing image as described above.

[0046] In addition, when the logic instructions in the above-mentioned memory are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0047] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the implementation steps of the method for color homogenization based on the original remote sensing image as described above.

[0048] In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the implementation steps of the method for color homogenization based on the original remote sensing image as described above.

[0049] Those skilled in the art may realize that, in combination with the embodiments disclosed in this application, it can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to functionality in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not exceed the scope of the present invention. The steps of the methods or algorithms described in the embodiments disclosed in this application can be directly implemented by hardware, software modules executed by a processor, or a combination of both. The software modules can be placed in a random access memory, internal memory, read-only memory, electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the technical field.

[0050] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A color homogenization method based on original remote sensing images, characterized in that, It includes the following steps: Step1. Geometrically correct the original remote sensing image to obtain the corrected image; Step2. Resample the corrected image according to the target resolution; Step3. Perform preliminary color homogenization on the resampled image, mosaic the color homogenization result to generate a template image; Step4. Divide the original remote sensing image into blocks according to a preset size and overlap degree, and record the starting point, width, and height information of each block image; Step5. Calculate the corresponding longitude and latitude regions of the block image in the template image; Step6. Calculate the mean and variance of the corresponding longitude and latitude regions in the block image and the template image. According to the calculated mean and variance, use the bilinear interpolation method to interpolate the mean and variance of each pixel, and obtain a mean image and a variance image of the same size as the original remote sensing image. Perform the same operation on the template image to obtain a reference mean image and a reference variance image of the same size as the original remote sensing image; Step7. Perform fine color homogenization on the original remote sensing image pixel by pixel in combination with the interpolated mean image, variance image, reference mean image, and variance image.

2. The uniform color method based on the original remote sensing image according to claim 1, wherein The geometric correction of the original remote sensing image includes: Perform vegetation enhancement on each original remote sensing image, and use rational polynomial coefficients to geometrically correct the original remote sensing image to obtain the corrected image. The corresponding coordinates of the corrected image are geographic coordinates.

3. A color homogenization method based on original remote sensing images according to claim 1, characterized in that The resampling of the corrected image according to the target resolution includes: Statistically analyze the resolution of the corrected image to obtain the minimum resolution. Multiply the minimum resolution by the coefficient N as the resampling target resolution, and resample the corrected image to make the resolutions of all corrected images consistent.

4. According to the color homogenization method based on the original remote sensing image described in claim 3, characterized in that The coefficient N is obtained by dividing the brightness mean of the minimum resolution image by the brightness variance of the minimum resolution image, and then multiplying by a preset value.

5. A color homogenization method based on original remote sensing images according to claim 1, characterized in that, The preliminary color homogenization of the resampled image and the mosaicking of the color homogenization result include: Unify the resolutions of all corrected images, perform preliminary color homogenization on the resampled image to reduce the color difference between all resampled images, and mosaic the preliminarily color homogenized image to generate a template image.

6. A color homogenization method based on original remote sensing images according to claim 1, characterized in that, The calculation of the corresponding longitude and latitude regions of the block image in the template image includes: Use the rational polynomial coefficients of the original remote sensing image data to convert the starting point of each block image from pixel coordinates to geographic coordinates, and obtain the corresponding region of the block image on the template image according to the converted geographic coordinates.

7. A color homogenization method based on the original remote sensing image according to claim 1, characterized in that, The calculation of the mean and variance of the corresponding longitude and latitude regions in the block image and the template image includes: Statistically analyze the pixel values of the block image and the block reference image by band respectively. According to the pixel values, calculate the mean and variance of each band to obtain the corresponding variance image and mean image.

8. A color homogenization method based on the original remote sensing image according to claim 1, characterized in that, The fine color homogenization of the original remote sensing image pixel by pixel includes: Taking the template image as the reference target, traverse the pixels of the original remote sensing image by band, use the rational polynomial coefficients to convert the original remote sensing image from pixel coordinates to geographic coordinates, and obtain the corresponding position of each pixel coordinate in the original remote sensing image in the template image based on the geographic coordinates.

9. A color homogenization system based on original remote sensing images, characterized in that, It includes: An image loading module, which is used to read and parse remote sensing data to obtain basic information such as rational polynomial coefficients, image width and height, and the starting point of the image; A geometric correction module, which uses the obtained rational polynomial coefficients and publicly provided elevation information to perform geometric correction on the image, converting the coordinates of the original remote sensing image from pixel coordinates to geographic coordinates; A vegetation enhancement module, which processes images including the near-infrared band and the red band to enhance the color of the vegetation area; A resampling module, which performs resolution resampling on the input remote sensing to make the resolution of the input remote sensing data consistent; An equalization module, which uses the input reference image or band reference data value to perform equalization on the original remote sensing data and outputs the equalization result.

10. An electronic device, characterized in that, It includes a memory and a processor, and the processor is used to implement a method for equalizing the original remote sensing image according to any one of claims 1 to 8 when executing the computer program stored in the memory.

Images