Remote sensing image enhancement method and system based on homomorphic filtering topographic correction

Abstract

U I T T R E K S E L Disclosed is a remote sensing image enhancement method and system based on homomorphic filtering topographic correction. The method includes: after acquiring a remote sensing image to be enhanced, converting the image from a red—green—blue (RGB) color space to a hue—saturation—value (HSV) color space, and extracting a brightness component; performing homomorphic filtering processing on the brightness component; combining’ the processed. brightness component with saturation and. hue, and. inverting the component back to the RGB color space to complete image enhancement; acquiring grayscale distribution of the enhanced image, and calculating image information entropy; further calculating contrast of the image through grayscale values of pixels; calculating an enhancement score of the image based on the information entropy' and. contrast, and. comparing' the enhancement score with a preset score to determine whether the image enhancement is completed. The present invention can effectively improve the image quality of remote sensing images. (+ Fig. 1)

Description

P2262 / NL REMOTE SENSING IMAGE ENHANCEMENT METHOD AND SYSTEM BASED ON HOMOMORPHIC FILTERING TOPOGRAPHIC CORRECTION TECHNICAL FIELD The present invention relates to the technical field of re mote sensing image enhancement, and in particular to a remote sensing image enhancement method and system based on homomorphic filtering topographic correction. BACKGROUND The quality of remote sensing images directly affects the ac curacy of image analysis and interpretation; therefore, image en hancement holds significant application value in remote sensing image processing. Remote sensing images are often affected by im aging conditions, sensor characteristics, as well as the atmos phere and terrain, resulting in issues such as noise, low con trast, and loss of details in the images. Although traditional im age enhancement methods like histogram equalization and contrast stretching can improve the overall quality of images, they often fail to effectively enhance details in different image regions and tend to introduce artifacts in highnoise environments, which af fects the accuracy of the images. Homomorphic filtering is a method that uses frequencydomain filtering to enhance specific frequency components of an image and is widely applied in enhancing the brightness and contrast of im- ages. By separating and processing the lowfrequency and high frequency components of an image, homomorphic filtering can en hance the detailed information of the image, and it is particular ly suitable for remote sensing images with uneven illumination or low contrast. However, traditional homomorphic filtering methods often rely on manually set filtering parameters, and during the processing, problems such as overenhancement or underenhancement of the image may occur, lacking automation and accuracy. Therefore, there is an urgent need to invent a remote sensing image enhancement technology to solve the problem in the existing technology that when remote sensing images are enhanced through homomorphic filtering, the terrain influence cannot be effectively eliminated, thereby resulting in low image quality. SUMMARY In view of this, the present invention provides a remote sensing image enhancement method and system based on homomorphic filtering topographic correction, aiming to solve the problem in the current technology that when remote sensing images are en hanced through homomorphic filtering, the terrain influence cannot be effectively eliminated, thereby leading to low image quality. The present invention provides a remote sensing image en hancement method based on homomorphic filtering topographic cor rection, including the following steps of: acquiring a remote sensing image to be enhanced, converting the remote sensing image from a redgreenblue (RGB) color space to a huesaturationvalue (HSV) color space, and extracting a brightness component from the remote sensing image; performing homomorphic filtering processing on the brightness component based on a high-pass filter, enhancing highfrequency components in the brightness component, and reducing low-frequency components in the brightness component; combining the brightness component after homomorphic filter ing processing with saturation and hue, and inverting the compo nent back to the RGB color space to realize image enhancement of the remote sensing image; acquiring image grayscale distribution in the remote sensing image after image enhancement, and determining information entropy of the remote sensing image after image enhancement according to the image grayscale distribution; acquiring grayscale values of pixels in the remote sensing image after image enhancement, and determining contrast of the re mote sensing image after image enhancement based on the grayscale values; and determining an enhancement score of the remote sensing image after image enhancement according to the information entropy and contrast, and comparing the enhancement score with a preset en hancement score, where: if the enhancement score is greater than or equal to the pre set enhancement score, it is determined that the image enhancement of the remote sensing image is completed; and if the enhancement score is lower than the preset enhancement score, it is determined that the image enhancement of the remote sensing image is not completed, and the remote sensing image is subjected to homomorphic filtering processing again until the en hancement score is greater than or equal to the preset enhancement score. Further, during the performing homomorphic filtering pro cessing on the brightness component based on a highpass filter, the method includes: preconfiguring a preset highfrequency gain and a preset low frequency gain, and configuring the highpass filter according to the preset highfrequency gain to perform homomorphic filtering processing on the brightness component, where an expression of ho momorphic filtering processing is as follows: H(u , v)-(R hl R)<1 e (%)2) o -+Rl where H(uJÛ is the brightness component after homomorphic filtering processing, u is the highfrequency component, and v is the lowfrequency component; Rh is the preset highfrequency gain, R, is the preset lowfrequency gain, D(uJÙ is a frequency of a current frequency point, DO is a lowfrequency cutoff frequency, and C is a preset amplitude of gain change. Further, after the performing homomorphic filtering pro cessing on the brightness component based on a highpass filter, the method includes: dividing the brightness component after homomorphic filtering processing into regions according to a preset distance, and per forming histogram equalization processing on the divided regional images; acquiring the total number of pixel points in each of the re gional images and a grayscale level of each pixel point, and de termining the total number of grayscale levels in the regional im age according to the total number of pixel points and the gray scale level of each pixel point; determining a grayscale value threshold of the region accord ing to the total number of pixel points and the total number of grayscale levels in the regional image; and acquiring the total number of grayscale levels of each re gional image after histogram equalization processing, and deter mining whether to adjust the preset distance according to the re lationship between the total number of grayscale levels and the grayscale value threshold, where: if the total number of grayscale levels is less than or equal to the grayscale value threshold, it is determined that the preset distance is not adjusted; and if the total number of grayscale levels is higher than the grayscale value threshold, it is determined that the preset dis tance is adjusted. Further, during the performing histogram equalization pro cessing on the regional images, the method includes: performing grayscale transformation on the regional image through cumulative distribution functions (CDFs); acquiring a CDF of each grayscale level, and extracting the cumulative distribution probability of each grayscale level; substituting cumulative distribution probability of each grayscale level into Formula I to obtain a preset grayscale range, and mapping an original grayscale value to a preset grayscale range to ensure that the distributed grayscale value covers the preset grayscale range, where Formula I is as follows: S(k) : CDF(k) CDFmin 1-CDFm where CDF(k) is a cumulative distribution value of a grayscale value k, CDFmm is a nonzero minimum cumulative distribution val ue, and S(k) is a preset grayscale distribution range. Further, during the performing homomorphic filtering pro cessing on the brightness component based on a highpass filter, the method further includes: performing multiscale decomposition on the brightness compo nent based on a wavelet transform formula to decompose the remote sensing image into a lowfrequency part and a highfrequency part, where the wavelet transform formula is as follows: 1 tb Zub) = Z() VE a where Zmb) represents a lowfrequency part and high frequency part of the decomposed remote sensing image, a is a scaling factor, b is a translation factor, and t is the brightness component; and the lowfrequency part contains overall illumination and structural information of the remote sensing image, and the high frequency part contains texture and grayscale information of the remote sensing image. Further, during the acquiring image grayscale distribution in the remote sensing image after image enhancement, and determining information entropy of the remote sensing image after image en hancement according to the image grayscale distribution, the meth od includes: acquiring occurrence probability of each grayscale value in the remote sensing image, and substituting the occurrence proba bility of each grayscale value into Formula II to obtain the in formation entropy of the remote sensing image, where Formula II is as follows: n H = Zpi10gpi i=1 where H is information entropy of the remote sensing image, m is occurrence probability of a grayscale value i, and n is the total number of grayscale levels. Further, during the acquiring grayscale values of pixels in the remote sensing image after image enhancement, and determining contrast of the remote sensing image after image enhancement based on the grayscale values, the method includes: acquiring a width and length of the remote sensing image, the grayscale value of each pixel point in the remote sensing image, and an average grayscale value of the remote sensing image; substituting the width and length of the remote sensing im age, the grayscale value of each pixel point in the remote sensing image, and the average grayscale value of the remote sensing image into Formula III to obtain the contrast of the remote sensing im age, where Formula III is as follows: 1 M .. _2 c= m;;(mf} where C is contrast of the remote sensing image, f(ùj) is a grayscale value of the pixel point, f is an average grayscale val ue of the remote sensing image, R4 is a width of the remote sensing image, and N is a length of the remote sensing image. Further, during the determining an enhancement score of the remote sensing image after image enhancement according to the in formation entropy and contrast, the method includes: acquiring an information entropy difference between the in formation entropy and a preconfigured preset information entropy, and determining the enhancement score according to the relation ship among the information entropy difference, a first preset in formation entropy difference and a second preset information en tropy difference: if the information entropy difference is less than the first preset information entropy difference, and the information entropy difference is greater than zero, the enhancement score being de termined to be Wl; if the information entropy difference is greater than or equal to the first preset information entropy difference, and the information entropy difference is less than the second preset in formation entropy difference, the enhancement score being deter- mined to be W2; if the information entropy difference is greater than or equal to the second preset information entropy difference, the en hancement score being determined to be W3; where the first preset information entropy difference is less than the second preset information entropy difference, and the first preset information entropy difference is greater than zero, Wl < W2 < W3. Further, during determining the enhancement score as Wi (i=1, 2, 3), the method includes: acquiring a contrast difference between the contrast and a preset contrast, and determining whether to adjust the enhancement score Wi according to a relationship between the contrast differ ence, and a preconfigured first preset contrast difference and second preset contrast difference; if the contrast difference is less than the first preset con trast difference, it being determined that the enhancement score Wi is not adjusted; if the contrast difference is greater than or equal to the first preset contrast difference and the contrast difference is less than the second preset contrast difference, an adjustment co efficient being determined to be gl, and the enhancement score Wi being adjusted according to an adjustment coefficient gl; if the contrast difference is greater than or equal to the second preset contrast difference, an adjustment coefficient being determined to be g2, and the enhancement score Wi being adjusted according to an adjustment coefficient g2; where the first preset contrast difference is less than the second preset contrast difference, gl<ig2<ld Compared with the existing technology, the beneficial effects of the present invention are as follows: through the topographic correction method based on homomorphic filtering, the brightness and contrast of remote sensing images can be significantly im proved, and it is especially suitable for remote sensing images with lost details due to uneven illumination or low contrast. In remote sensing image processing, uneven illumination and exces- sively low contrast often affect the quality of images, reduce the visibility of image details, and thus affect the accuracy of image analysis and interpretation. By converting the remote sensing im age from the RGB color space to the HSV color space, extracting the brightness component and performing highpass filtering on the component, the highfrequency components in the image can be en hanced, thereby improving the presentation of image details. This method effectively avoids the situations of overenhancement or underenhancement of contrast in traditional image enhancement methods, and it has important practical significance especially in the application scenarios of remote sensing images. In addition, by automatically calculating the grayscale distribution, infor mation entropy, and contrast of the image, it is ensured that the effect of image enhancement meets the expected standards. The in formation entropy can reflect the complexity of an image, while contrast directly affects the visual effect of the image. Through the quantitative evaluation of these two indicators, it is possi ble to objectively determine whether the effect of the enhanced image meets the ideal level. The automatic contrast evaluation during the enhancement process avoids the inconsistency in effects that may occur when manually adjusting parameters in the tradi tional method, and improves the automation and accuracy of image enhancement. This automatic adjustment mechanism based on stand ardized evaluation makes remote sensing image enhancement more in telligent and efficient, with stronger adaptability, and can meet the needs of different application scenarios. Finally, by setting a preset enhancement score and comparing it with the actual en hancement score, the quality control of image enhancement is en sured. If the enhancement effect does not meet the expectation, the system will automatically start the iterative process of homo- morphic filtering until the enhancement effect of the image meets the standard. This adaptive adjustment mechanism avoids the cum- bersomeness of manual operations, can perform precise optimization according to the characteristics of different images, and improves the robustness and stability of image enhancement. This innovative measure not only improves the automation level of image processing but also enables the enhanced images to provide higher accuracy in subsequent analysis, meeting the high-precision requirements of remote sensing data analysis, change detection, and other fields. On the other hand, the present application also provides a remote sensing image enhancement system based on homomorphic fil tering topographic correction, including: an acquisition module, configured to acquire a remote sensing image to be enhanced; an extraction module, electrically connected to the acquisi tion module, and configured to convert the remote sensing image from an RGB color space to an HSV color space, and extract a brightness component from the remote sensing image; a central control module, electrically connected to the ex traction module, and configured with a highpass filter, to per form homomorphic filtering processing on the brightness component based on a highpass filter, enhance highfrequency components in the brightness component, and reduce lowfrequency components in the brightness component; and to combine the brightness component after homomorphic filtering processing with saturation and hue, and invert the component back to the RGB color space to realize image enhancement of the remote sensing image; and an evaluation module, electrically connected to the central control module, and configured to acquire image grayscale distri bution in the remote sensing image after image enhancement, and determine information entropy of the remote sensing image after image enhancement according to the image grayscale distribution; to acquire grayscale values of pixels in the remote sensing image after image enhancement, and determine contrast of the remote sensing image after image enhancement based on the grayscale val ues; and to determine an enhancement score of the remote sensing image after image enhancement according to the information entropy and contrast, and comparing the enhancement score with a preset enhancement score, and determine whether the image enhancement of the remote sensing image is completed according to a relationship between the enhancement score and the preset enhancement score. It can be understood that the remote sensing image enhance ment method and system based on homomorphic filtering topographic correction in the above-mentioned embodiments of the present in- vention have the same beneficial effects, which will not be re peated here. BRIEF DESCRIPTION OF DRAWINGS Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following de tailed description of preferred embodiments. The accompanying drawings are for the purpose of illustrating preferred embodiments only and are not to be construed as limiting the present inven tion. Furthermore, like reference numerals are used throughout the drawings to designate like parts. In the drawings: FIG. 1 is a flow chart of a remote sensing image enhancement method based on homomorphic filtering topographic correction pro vided in an embodiment of the present invention; FIG. 2 is an effect diagram of converting a remote sensing image from an RGB color space to an HSV space in an embodiment of the present invention; FIG. 3 is an effect diagram of homomorphic filtering remote sensing image topographic enhancement based on wavelet transform in an embodiment of the present invention; and FIG. 4 is a functional block diagram of a remote sensing im age enhancement system based on homomorphic filtering topographic correction provided in an embodiment of the present invention. DETAILED DESCRIPTION OF EMBODIMENTS Exemplary embodiments of the present disclosure will be de scribed in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be implemented in various forms and is not to be limited by the embodiments set forth herein. Rather, these embodi- ments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. In the case of no con f1ict, the embodiments in the present invention and the features in the embodiments can be combined with each other. Hereinafter, the present invention will be described in detail with reference to the accompanying drawings in conjunction with embodiments. As shown in FIGS. 1 to 3, in some embodiments of the present application, the present embodiment provides a remote sensing im age enhancement method based on homomorphic filtering topographic correction, including that: In step 5100: a remote sensing image to be enhanced is ac quired, the remote sensing image is converted from an RGB color space to an HSV color space, and a brightness component is ex tracted from the remote sensing image. It can be understood that by converting the remote sensing image from the RGB color space to the HSV color space, a luminance component (V component) in the HSV space is utilized to represent the luminance information of the image. The RGB color space is mainly used to represent colors, while the HSV color space is more suitable for processing image features related to brightness, sat uration and hue. The converted luminance component (V component) can independently reflect the luminance characteristics of the im age, which is convenient for subsequent image enhancement pro cessing. By extracting the luminance component, the luminance and color information of the image can be effectively separated. Specifically, during the performing homomorphic filtering processing on the brightness component based on a highpass fil ter, the method further includes that: multiscale decomposition is performed on the brightness component based on a wavelet trans form formula to decompose the remote sensing image into a low freguency part and a highfrequency part, where the wavelet trans form formula is as follows: Zaß(Ü==j%Z(%?), where Zab(Ü represents a lowfrequency part and highfrequency part of the decomposed re mote sensing image, a is a scaling factor, b is a translation fac tor, and t is the brightness component; and the lowfrequency part contains overall illumination and structural information of the remote sensing image, and the highfrequency part contains texture and grayscale information of the remote sensing image. It can be understood that based on the wavelet transform for mula, by adjusting a scaling factor (a) and a translation factor (b), the image can be analyzed at different scales and positions. Wavelet transform can capture the detail changes of images at dif ferent resolutions, and has strong adaptability, which can effec tively deal with the separation of different frequency components in images. Specifically, the lowfrequency part represents the overall illumination information and structural information of the remote sensing image, including the macroscopic features of the image, such as largescale illumination changes, the overall structure and shape of the image, etc. The high frequency part contains the detailed information in the image, such as texture, edge and gray change, which reflects the microscopic features of the image. By decomposing the luminance components of remote sens ing images into lowfrequency and highfrequency parts, the system can perform different enhancement processing for different fre quency components. The processing of the lowfrequency part usual ly aims to enhance the illumination and overall structure of the image, and improve the uniformity and clarity of the image. The processing of highfrequency parts can focus on enhancing the tex ture and details of the image, and improve the sharpness and con trast of the image. Through this separation process, all aspects of the image can be adjusted more accurately, the information loss in the overall image processing can be avoided, and the visual ef fect of the image can be improved. In addition, the multiscale nature of wavelet transform enables it to analyze images at dif ferent scales, which provides great advantages for processing com plex remote sensing images. Remote sensing images often contain multilevel and multiscale features, such as macroscopic struc ture of terrain and tiny details. Wavelet transform can extract the relevant information of images on different scales, so that the image can better reflect the overall shape and local details of ground objects after processing. This multi-scale processing method is especially suitable for the analysis of remote sensing images, because remote sensing images are usually affected by fac- tors such as illumination changes and viewing angle changes. Wave let transform can effectively reduce the interference of these factors and ensure the accurate extraction of important features in the images. Another advantage of wavelet transform is its lo calization property in time and frequency. Unlike traditional Fou- rier transform, Fourier transform is global and cannot provide in formation about local features. Wavelet transform can effectively capture frequency changes in local areas, especially suitable for processing images with local texture and detail changes. In remote sensing image processing, local features such as buildings, roads, vegetation and other details often determine the practical appli cation value of images. Wavelet transform can accurately process these local areas and enhance their detail performance. In step SZOO, homomorphic filtering processing is performed on the brightness component based on a highpass filter, high frequency components in the brightness component are enhanced, and lowfrequency components in the brightness component is reduced. Specifically, during the homomorphic filtering processing be ing performed on the brightness component based on a highpass filter, the method includes: a preset highfrequency gain and a preset lowfrequency gain are preconfigured, and the highpass filter is configured according to the preset highfrequency gain to perform homomorphic filtering processing on the brightness com ponent, where an expression of homomorphic filtering processing is as follows: H(u , v)(R h R)<1 e (%)2) o -+Rl where H(uJÛ is the brightness component after homomorphic filtering processing, u is the highfrequency component, and v is the lowfrequency component; Rh is the preset highfrequency gain, R, is the preset lowfrequency gain, D(uJÙ is a frequency of a current frequency point, DO is a lowfrequency cutoff frequency, and c is a preset amplitude of gain change. It can be understood that by presetting the highfrequency gain (Rh) and the lowfrequency gain (R,), and configuring the highpass filter to perform homomorphic filtering on the luminance component, it is possible to enhance the highfrequency component (detailed portion) in the image while suppressing the low frequency component (flat area or background information). The core idea of homomorphic filtering is to decompose the image into lowfrequency and highfrequency components through frequency do main transformation, and then adjust them separately: the high frequency part is enhanced (Rh gain), and the lowfrequency part is attenuated (Rl gain) to improve the contrast and detail perfor mance of the image. By setting the lowfrequency cutoff frequency (Do) and the gain change amplitude (c), the influence range and intensity of filtering can be accurately controlled, thereby achieving optimal image processing. This method is especially suitable for improving the clarity of details in remote sensing images, while reducing the interference of background noise. Specifically, after the homomorphic filtering processing be ing performed on the brightness component based on a highpass filter, the method includes that: the brightness component is di vided after homomorphic filtering processing into regions accord ing to a preset distance, and histogram equalization processing is performed on the divided regional images; the total number of pix el points in each of the regional images and a grayscale level of each pixel point are acquired, and the total number of grayscale levels in the regional image are determined according to the total number of pixel points and the grayscale level of each pixel point; a grayscale value threshold of the region is determined ac cording to the total number of pixel points and the total number of grayscale levels in the regional image; and the total number of grayscale levels of each regional image is acquired after histo gram equalization processing, and it is determined whether to ad just the preset distance according to the relationship between the total number of grayscale levels and the grayscale value thresh old, where: if the total number of grayscale levels is less than or equal to the grayscale value threshold, it is determined that the preset distance is not adjusted; and if the total number of grayscale levels is higher than the grayscale value threshold, it is determined that the preset distance is adjusted. Specifically, during the histogram equalization processing being performed on the regional images, the method includes that: grayscale transformation being performed on the regional image through CDFs; a CDF of each grayscale level is acquired, and the cumulative distribution probability of each grayscale level is ex tracted; cumulative distribution probability of each grayscale level is substituted into Formula I to obtain a preset grayscale range, and an original grayscale value is mapped to a preset gray scale range to ensure that the distributed grayscale value covers the preset grayscale range, where Formula I is as follows: S(k)= ÊEËËËËËËËË, where CDF(k) is a cumulative distribution value of a grayscale value k, CDFmm is a nonzero minimum cumulative distri bution value, and S(k) is a preset grayscale distribution range. It can be understood that after the luminance component is subjected to the homomorphic filtering process, the image is di vided into a plurality of small regions by region division, so that individual image enhancement processing is performed for each region. This method can not only equalize the brightness of the whole image, but also adjust the brightness difference of local areas, thus enhancing the details of the image, especially in com plex terrain and lighting conditions, where local features of the image may be concealed. Through area division, the image pro cessing effect of each area can be controlled more accurately, so that the brightness difference of the image between different are as can be effectively improved. The histogram equalization process calculates the CDF of each area image, and uses the histogram to gray transform the gray value of the image. The CDF reflects the pixel distribution of each gray level in the image, and is usually used to evaluate the contrast of the image. If the gray level dis tribution of the image is concentrated in a small range, the con trast of the image is low and the details are difficult to identi fy. Through histogram equalization, these gray values can be even ly distributed in the whole gray range, enhancing the contrast and detail level of the image, and making the brightness and details of the image more clearly visible. To ensure that the effect of image enhancement meets the expectation, the gray transformation of each area should be mapped not only according to the CDF, but also based on the preset gray range, so as to avoid excessive or insufficient brightness adjustment. In addition, this method fur ther optimizes the adaptive ability of image processing by dynami cally adjusting the preset distance of region division. When the gray level distribution of a certain area is lower than the preset threshold, the system automatically identifies and decides whether to adjust the division distance of the area to ensure the best processing effect of the image in different areas. This adaptive adjustment mechanism makes the image processing process more in telligent, and can flexibly adjust the processing strategy accord ing to the actual characteristics of the image, thus avoiding the shortcomings of manual intervention and static parameter setting. By adjusting the preset distance, the system can perform finer processing on areas with uneven gray level distribution, ensuring that the brightness distribution of the entire image is more uni form, thus improving the quality and readability of the image. Fi nally, the gray transformation formula of histogram equalization maps the original gray values by using the minimum value of CDF (CDFmm) and the preset gray distribution range (S(k)), thus ensur ing that the distributed gray values cover the preset gray range. Through this mapping, the gray value of the image will be more uniformly distributed in the whole gray space, thus greatly im proving the overall contrast of the image. The CDF(k) in the formu la represents the cumulative distribution value of the current gray level, while by comparison with CDFmm, the system can identi fy the smallest gray value and map it to the appropriate gray range, preventing the image from appearing over bright or over dark areas. This process not only enhances the details of the im age, but also reduces the interference of noise to a certain ex tent and improves the effect of image processing. In step S300, the brightness component is combined after ho momorphic filtering processing with saturation and hue, and the component is inverted back to the RGB color space to realize image enhancement of the remote sensing image. It can be understood that in the homomorphic filtering pro cess, the luminance component is processed by a highpass filter to remove the lowfrequency portion and enhance the highfrequency portion, thereby improving the detail and contrast of the image. The processed luminance component is then combined with the satu ration and hue (i.e. color information) of the image to be trans ferred from the HSV color space back to the RGB color space by in verse transformation. This process enhances the visual effect of the image by reconstructing the relationship between brightness and color, especially in terms of illumination and detail presen tation, thus improving the overall quality of remote sensing imag es. Through this technology, remote sensing images with low con trast or uneven illumination can be effectively improved, the vis ual performance can be enhanced, and the images can be clearer and more recognizable. In step S400: image grayscale distribution in the remote sensing image is acquired after image enhancement, and information entropy of the remote sensing image is determined after image en hancement according to the image grayscale distribution. Specifically, during the image grayscale distribution in the remote sensing image being acquired after image enhancement, and information entropy of the remote sensing image being determined after image enhancement according to the image grayscale distribu tion, the method includes that: occurrence probability of each grayscale value in the remote sensing image is acquired, and the occurrence probability of each grayscale value is substituted into Formula II to obtain the information entropy of the remote sensing image, where Formula II is as follows: II=-§X;1pogpi, where H is information entropy of the remote sensing image, m is occurrence probability of a grayscale value i, and n is the total number of grayscale levels. It can be understood that the gray distribution of the remote sensing image after image enhancement is obtained by counting the frequency of occurrence of pixels of each gray level in the image. The probability of occurrence (m) of each gray value reflects the proportion of this gray level in the image. Then, by substituting these probabilities into the information entropy Formula H, the overall information entropy of the image is calculated. As an im portant concept in information theory, information entropy is a tool to measure image complexity and information quantity. The higher the value, the greater the amount of information and the richer the details of the image, otherwise, the image may be mo notonous or lack details. The calculation process of information entropy is actually an indepth analysis of image gray value dis tribution. It can reflect the diversity and complexity of images by considering the distribution of each gray level in the image. Specifically, images with higher information entropy usually show richer texture, details and higher recognizability. For example, in remote sensing images, details such as land types, buildings, vegetation, etc., usually require higher contrast and detail per formance. Images with higher information entropy can better pre sent these detailed information, having stronger image expressive ness. Through the evaluation of information entropy, it can pro vide objective standards for image enhancement and help determine whether the effect of image enhancement achieves the expected goal. This process can not only be used to evaluate the quality of remote sensing images, but also be widely used in other image pro cessing fields, such as medical images and satellite images. In remote sensing image processing, problems such as uneven illumina tion and low contrast are often faced. As an objective quantita tive index, information entropy can effectively help to judge whether the image is improved and ensure the quality of the en hancement effect. In step S500: grayscale values of pixels are acquired in the remote sensing image after image enhancement, and contrast of the remote sensing image is determined after image enhancement based on the grayscale values. Specifically, during the grayscale values of pixels in the remote sensing image being acquired after image enhancement, and contrast of the remote sensing image being determined after image enhancement based on the grayscale values, the method includes: a width and length of the remote sensing image, the grayscale value of each pixel point in the remote sensing image, and an average grayscale value of the remote sensing image are acquired; the width and length of the remote sensing image, the grayscale value of each pixel point in the remote sensing image, and the average grayscale value of the remote sensing image are substituted into Formula III to obtain the contrast of the remote sensing image, where Formula III is as follows: C: ,ÈZïlzy=1U(i,j)f)2, where C is contrast of the remote sensing image, f(Lj) is a grayscale value of the pixel point, f is an average grayscale value of the remote sensing image, A4 is a width of the remote sensing image, and N is a length of the remote sensing image. It can be understood that the contrast of the image is achieved by calculating the difference between the gray value of each pixel and the average gray value of the image. The gray value of each pixel in the image reflects the brightness information of that pixel, while the average gray value of the image represents the overall brightness level of the image. By combining these gray values with the size information of the image, the contrast of the image can be quantified. Each item in Formula III represents an important factor to be considered in the process of contrast cal culation. For example, a gray value f(ÿD of the pixel points re flects the brightness of each position, and the average gray value of the image is used as a reference value to measure the degree of deviation of each pixel from the overall brightness. M and N rep resent the width and length of the image, respectively, which pro vide the size information of the image for the calculation pro cess, so that the calculation of contrast can take into account the overall scale of the image. The contrast value calculated in this way not only provides a quantitative basis for image quality evaluation, but also helps to optimize the process of image en hancement. The higher the contrast ratio means the more detailed the image is and the clearer the visual effect. This is particu larly important in the processing of remote sensing images, be cause remote sensing images usually have a complex background of ground features, and low contrast images will lead to blurring or confusion of different ground features. Therefore, enhancing image contrast can effectively improve the visual effect of the image, make the features of ground features more prominent, and facili tate subsequent analysis and application. In addition, the result of contrast calculation can also be used as an evaluation standard for whether the image enhancement achieves the expected effect. If the contrast value of the image is low, it may mean that the image enhancement processing is insufficient and further processing or adjustment may be required. However, if the contrast value is high, it indicates that the details in the image have been effec tively enhanced and the quality of the image has been improved. In this process, using contrast as the standard of image quality evaluation can not only achieve more accurate image processing, but also reduce manual intervention and improve the automation and efficiency of remote sensing image enhancement processing. In step S600, an enhancement score of the remote sensing im age is determined after image enhancement according to the infor mation entropy and contrast, and the enhancement score is compared with a preset enhancement score. Specifically, during determination of the enhancement score of the remote sensing image after image enhancement according to the information entropy and contrast, the method includes that: an information entropy difference between the information entropy and a preconfigured preset information entropy is acquired, and the enhancement score is determined according to the relationship among the information entropy difference, a first preset infor mation entropy difference and a second preset information entropy difference: if the information entropy difference is less than the first preset information entropy difference, and the information entropy difference is greater than zero, the enhancement score is determined to be W1; if the information entropy difference is greater than or equal to the first preset information entropy dif ference, and the information entropy difference is less than the second preset information entropy difference, the enhancement score is determined to be W2; if the information entropy differ ence is greater than or equal to the second preset information en tropy difference, the enhancement score is determined to be W3; where the first preset information entropy difference is less than the second preset information entropy difference, and the first preset information entropy difference is greater than zero, W1 < W2 < W3. Specifically, during determination of the enhancement score as Wi (i=1, 2, 3), the method includes that: a contrast difference between the contrast and a preset contrast is acquired, and it is determined whether to adjust the enhancement score Wi according to a relationship between the contrast difference, and a preconfig ured first preset contrast difference and second preset contrast difference; if the contrast difference is less than the first pre set contrast difference, it is determined that the enhancement score Wi is not adjusted; if the contrast difference is greater than or equal to the first preset contrast difference and the con trast difference is less than the second preset contrast differ ence, an adjustment coefficient is determined to be gl, and the enhancement score Wi is adjusted according to an adjustment coef ficient g1; if the contrast difference is greater than or equal to the second preset contrast difference, an adjustment coefficient is determined to be g2, and the enhancement score Wi is adjusted according to an adjustment coefficient g2; where the first preset contrast difference is less than the second preset contrast dif ference, g1<ig2<:1. Specifically, when comparing between the enhanced score and the preset enhanced score, the step includes if the enhancement score is greater than or equal to the preset enhancement score, it is determined that the image enhancement of the remote sensing im age is completed; and if the enhancement score is lower than the preset enhancement score, it is determined that the image enhance ment of the remote sensing image is not completed, and the remote sensing image is subjected to homomorphic filtering processing again until the enhancement score is greater than or equal to the preset enhancement score. It can be understood that by calculating the enhancement score of the image, the quality of the image is quantified into three levels, W1, W2, and W3, which represent different image quality levels, respectively, based on the change of information entropy. This scoring mechanism ensures that the contrast and de tail information of the image are effectively improved during the enhancement process by comparing the difference of information en tropy. Specifically, when the information entropy difference is less than a certain preset threshold, the image is considered to be in a lower enhancement state, and if the information entropy difference is large, the image may be in a higher enhancement state, and then classified by scoring W1, W2, and W3. Furthermore, the role of contrast in image quality evaluation cannot be ig- nored. By calculating the difference of image contrast and compar- ing it with the preset contrast difference, the effect of image enhancement is further refined. When the contrast difference falls within a certain threshold range, the enhancement score is fine tuned according to the adjustment factor g1 or g2 to ensure that the image is visually sharper, and the details are fully revealed. Through this adjustment mechanism, it is possible to avoid the problem that the image is excessively enhanced or the effect is not obvious due to the contrast being too high or too low. On this basis, the image enhancement score is compared with the preset target enhancement score, and finally the completion degree of im age enhancement is determined. If the enhancement score does not meet the preset criteria, the image enhancement process continues until the quality requirements are met. This iterative process en sures that image enhancement can be continuously optimized, avoid ing excessive enhancement or insufficient enhancement in one pro cessing, and improving the accuracy of image processing. Through repeated adjustment and optimization, the highquality remote sensing image enhancement effect is finally realized. Through this comprehensive scoring mechanism, using the quantitative relation ship between information entropy and contrast, combined with mul tiple iterative adjustments, the contrast and detail performance of the image are significantly improved. This method not only ef fectively avoids the deficiency of artificial adjustment in tradi tional methods, but also ensures the quality stability of image enhancement in the process of automatic processing, and adapts to the needs of remote sensing image enhancement in different scenar ios. In the above embodiment, the topographic correction method based on homomorphic filtering can significantly improve the brightness and contrast of the remote sensing image, and is par- ticularly suitable for those remote sensing images in which de tails are lost due to uneven illumination or low contrast. In re mote sensing image processing, uneven illumination and excessively low contrast often affect the quality of images, reduce the visi bility of image details, and thus affect the accuracy of image analysis and interpretation. By converting the remote sensing im age from the RGB color space to the HSV color space, extracting the brightness component and performing highpass filtering on the component, the highfrequency components in the image can be en hanced, thereby improving the presentation of image details. This method effectively avoids the situations of overenhancement or underenhancement of contrast in traditional image enhancement methods, and it has important practical significance especially in the application scenarios of remote sensing images. In addition, by automatically calculating the grayscale distribution, infor mation entropy, and contrast of the image, it is ensured that the effect of image enhancement meets the expected standards. The in formation entropy can reflect the complexity of an image, while contrast directly affects the visual effect of the image. Through the quantitative evaluation of these two indicators, it is possi ble to objectively determine whether the effect of the enhanced image meets the ideal level. The automatic contrast evaluation during the enhancement process avoids the inconsistency in effects that may occur when manually adjusting parameters in the tradi tional method, and improves the automation and accuracy of image enhancement. This automatic adjustment mechanism based on stand ardized evaluation makes remote sensing image enhancement more in telligent and efficient, with stronger adaptability, and can meet the needs of different application scenarios. Finally, by setting a preset enhancement score and comparing it with the actual en hancement score, the quality control of image enhancement is en sured. If the enhancement effect does not meet the expectation, the system will automatically start the iterative process of homo morphic filtering until the enhancement effect of the image meets the standard. This adaptive adjustment mechanism avoids the cum bersomeness of manual Operations, can perform precise optimization according to the characteristics of different images, and improves the robustness and stability of image enhancement. This innovative measure not only improves the automation level of image processing but also enables the enhanced images to provide higher accuracy in subsequent analysis, meeting the highprecision requirements of remote sensing data analysis, change detection, and other fields. In another preferred implementation based on the above embod- iment, as shown in FIG. 4, the present embodiment provides a re- mote sensing image enhancement system based on homomorphic filter ing topographic correction, including an acquisition module, an extraction module, a central control module, and an evaluation module. Specifically, the acquisition module is configured to acquire a remote sensing image to be enhanced. The extraction module is electrically connected to the acquisition module, and configured to convert the remote sensing image from an RGB color space to an HSV color space, and extract a brightness component from the re mote sensing image. The central control module is electrically connected to the extraction module, and configured with a high pass filter, to perform homomorphic filtering processing on the brightness component based on a highpass filter, enhance high frequency components in the brightness component, and reduce low frequency components in the brightness component; and to combine the brightness component after homomorphic filtering processing with saturation and hue, and invert the component back to the RGB color space to realize image enhancement of the remote sensing im age. The evaluation module is electrically connected to the cen tral control module, and the evaluation module is configured to obtain the image gray distribution in the image enhanced remote sensing image and determine the information entropy of the image enhanced remote sensing image according to the image gray distri bution. The evaluation module is further configured to acquire the gray value of each pixel of the image enhancement remote sensing image, and determine the contrast of the image enhancement remote sensing image by the gray value of each pixel. The evaluation mod ule is further configured to determine an enhancement score of the remote sensing image after image enhancement according to the in- formation entropy and contrast, and determine whether the remote sensing image has completed image enhancement according to the re- lationship between the enhancement score and the preset enhance ment score. It can be understood that the remote sensing image enhance ment method and system based on homomorphic filtering topographic correction in the abovementioned embodiments of the present in- vention have the same beneficial effects, which will not be re peated here. It will be understood by those skilled in the art that embod iments of the present application may be provided as methods, sys tems, or computer program commodities. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining soft ware and hardware aspects. Furthermore, the present application may take the form of computer program articles implemented on one or more computerusable storage media (including, but not limited to, magnetic disk storage, compact disc readonly memory (CDROM), optical memory, etc.) containing computerusable program code therein. The present application is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program articles according to embodiments of the pre sent application. It is to be understood that each flow and / or block in the flowchart and / or block diagrams, as well as combina tions of the flow and / or blocks in the flowchart and / or block dia grams, may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded pro cessor, or other programmable data processing device to produce a machine, so that the instructions, which, when executed by a pro cessor of the computer or other programmable data processing de vice, produce an apparatus for implementing the functions speci fied in the flow or flows of the flow chart and / or the block or blocks of the block diagram. These computer program instructions may also be stored in a computerreadable memory capable of directing a computer or other programmable data processing apparatus to operate in a particular manner such that the instructions stored in the computerreadable memory produce an article of manufacture including instruction means for implementing the functions specified in a flow or pro cesses of the flowcharts and / or a block or blocks of the block di agrams. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computerimplemented process, such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one flow or more flows of the flowcharts and / or one block or more blocks of the block diagrams. Finally, it is to be noted that the above embodiments are merely used to illustrate the technical solution of the present invention without limiting it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific embodi ments of the present invention, and any modifications or equiva 5 lent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention. C O N C L U S I E S l. Method for improving remote sensing images based on of homomorphic filter terrain correction, characterized by the encompassing ten of: obtaining an improved remote sensing image, where the remote sensing image is converted from an RGB color space to an HSV color space, and extracting the brightness com component from the remote sensing image; performing a homomorphic filter operation on the bright ity component using a high-pass filter, where the high frequency component of the brightness component is strength and the lag-frequency component of the brightness component is weakened; combining the results obtained after homomorphic filter processing brightness component with the saturation and hue, and the back convert to RGB color space for image enhancement to output the remote sensing image; obtaining the grayscale distribution of the image in the remote sensing image after image enhancement, and determining the information entropy of the remote sensing image after image enhancement consumption based on the grayscale distribution of the image; obtaining the gray value of the pixel of the remote sensor image after image enhancement, and determining the con trast of the remote sensing image after image enhancement based on of the gray value of the pixel; determining the remote sensing improvement score image after image enhancement based on information entropy and the contrast, and comparing the improvement score with a preset improvement score, where: if the improvement score is greater than or equal to the previous set improvement score, it is determined that the remote sen singbeeld has completed the image enhancement; if the improvement score is lower than the preset ver improvement score, it is determined that the remote sensing image image enhancement has not been completed, and the remote sensing image again subjected to homomorphic filter processing, until the improvement score is greater than or equal to the preset set improvement score. 2. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 1, notice that, when performing the homomorphic filter processing of the brightness component using a high-pass filter, the following includes: pre-configuring a pre-configured high-frequency frequent gain and a preset low-frequency gain king, and configuring the high-pass filter based on the pre-configured high-frequency gain in order to perform homomorphic filter processing of the brightness component run, where the expression for the homomorphic filter processing is the following is displayed: mu, u) {Rh RMI e*tfDü"Wïf + R; where H(u,v) is the brightness component after the homomorphic filter processing, u is the high frequency component, v is the low frequency component; Rh is the preset high frequency gain, Rl the preset low frequency gain is, D(u,v) de fre frequency at the current frequency point is, Do the low frequency off cutting frequency is, and c is the preset amplitude of the strength variation is. 3. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 2, notes that when the homomorphic filter processing of the hell the unit component has been performed, this includes: dividing the brightness component after the homomorphic filter processing in areas according to a preset distance, and performing a histogram equalization processing on the thus divided area images; obtaining the total number of pixels in each of the areas images as well as the gray level of each of the pixels, and the poles of the total number of gray levels in the area image on based on the total number of pixels and the gray levels of the pixels; determining the gray value threshold of the area on based on the total number of pixels and the total number of grayscale levels in the area image; obtaining the total number of gray levels of each area image after histogram equalization processing, and determining, on based on the relationship between the total number of gray levels and the grayscale threshold, or the preset distance must be adjusted, provided that: when the total number of gray levels is less than or equal to the threshold value of gray value is determined that the predefined set distance is not adjusted; when the total number of gray levels is greater than the threshold value of gray value, it is determined that the preset offset position is being adjusted. 4. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 3, notice that when running the Histogram equalization processing of the area images, this includes: performing a grayscale transformation on the area image by means of a cumulative distribution histogram; obtaining the cumulative distribution histogram of each gray level and extracting the cumulative distribution probability of each of the gray levels; substituting the cumulative distribution probability of each gray level in formula I to achieve a preset gray value to obtain range, and assigning the original gray value to the preset grayscale range so that the ver divided grayscale values ​​cover the preset range, where Formula I is shown as follows: 5 i.. CDFUa) GDF { ) 1 _ CDF _ where CDF(k) is the cumulative distribution value of the gray value k, CDFn the nonzero minimum cumulative distribution value is, and S(k) represents the preset grayscale range close. 5. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 4, notice that, when performing the homomorphic filter processing of the brightness component based on the high-pass filter, also includes: performing a multiscale decomposition of the clear ity component by means of a wavelet transform formula, in order to decompose the remote sensing image into a low-frequency frequent component and a high-frequency component, where the wave let transformation formula is shown as follows: l lb z} Fz (_) Va II where the zab(t) low frequency component and the high frequency com component to form the representation of the decomposed remote sen singbeeld, where a denotes the scale factor, b the translation indicates factor, and t represents the brightness component; where the low frequency component represents the global illumination and structure natural information of the remote sensing image, and the high frequency component the texture and grayscale information of displays the remote sensing image. 6. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 1, characterized notes that when obtaining the grayscale distribution of the image of the remote sensing image after image enhancement and determining the information entropy of remote sensing image after image enhancement based on the grayscale distribution of the image, includes: obtaining the probability of occurrence of each gray value of the remote sensing image, and substituting the probabilities of occurrence of the respective gray matter in formula ii, in order to obtain the information entropy of the remote to obtain a sensing image, where formula II is expressed as follows displayed: . H_ZP£lÜgP£ i 1 ; where H is the information entropy of the remote sensing image present, Pi the probability of occurrence of gray represents the value i, and n represents the total number of gray levels. 7. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 1, notices that when obtaining the gray value of the pixel of the remote sensing image after image enhancement and determination of the contrast of the remote sensing image after image enhancement ring based on the respective gray values ​​of the pixels, to barrel: obtaining the width and length of the remote sensing image, the grayscale values ​​of the respective pixels in the remote sensing image, as well as the average gray value of the remote sensing image; substituting the latitude and longitude of the remote sen image, the gray values ​​of the respective pixels and the average divide the gray value of the remote sensing image into formula iii, in order to obtain the contrast of the remote sensing image, where formula III is represented as follows: l MN .. _2 0 ;;(f(1=J}n where C represents the contrast of the remote sensing image close, f(i,j) indicates the gray value of the pixel, the average represents the grayscale value of the remote sensing image, M the width of the remote sensing image, and N is the length of the re mote sensing image displays. 8. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 1, notes that when determining the improvement score of the remote sensing image after image enhancement based on the information matieentropy and the contrast, includes: obtaining the difference in information entropy between the information entropy and the preconfigured preset information entropy, and determining the improvement score on basis of the relationship between this difference in information entropy and an initial preset information entropy difference and a second preset information entropy difference: when the difference in information entropy is less than the first preset information entropy difference and is greater then zero, the improvement score is determined as Wl; when the difference in information entropy is greater than or equal to the first preset information entropy difference and is less than the second preset information entropy difference, the improvement score is determined as W2; when the difference in information entropy is greater than or equal to the second preset information entropy difference, the improvement score is set as W3; where the first preset information entropy difference is less than the second preset information entropy difference and is greater than zero; and where W1 < W2 < W3. 9. Method for improving remote sensing images based on of homomorphic filter terrain correction according to claim 8, notes that when the improvement score is determined as Wi, where i = 1, 2, 3, includes the following: obtaining the difference between the contrast and a predetermined preset contrast value, and determining, on basis of the relationship between this contrast difference and a first preset contrast difference and a second preset contrast difference, or the Wi improvement score should be adjusted past; when the contrast difference is smaller than the first predefined stated contrast difference, it is determined that the improvement score Wi is not adjusted; when the contrast difference is greater than or equal to the front ste preset contrast difference and is less than the two the preset contrast difference, the adjustment coefficient is cient gl is determined, and the improvement score Wi is adjusted based on the adjustment coefficient gl; when the contrast difference is greater than or equal to the two the preset contrast difference, the adjustment coefficient is cient g2 is determined, and the improvement score Wi is adjusted based on the adjustment coefficient g2; where the first preset contrast difference is smaller then the second preset contrast difference, and gl < g2 < 1. 10. System for improving remote sensing images on ba sis of homomorphic filter terrain correction, using the method according to any of claims 19, characterized in that it includes: an acquisition module, configured to improve Remote to obtain a sensing image; an extraction module, electrically connected to the acquisition module, where the extraction module is configured to extract the Remote sen- convert image from RGB color space to HSV color space and the brightness component of the Remote sensing to extract image; a central control module, electrically connected to the extractor tie module, where the central control module is configured with a high-pass filter; the central control module performs, on basis of the high-pass filter, a homomorphic filter processing on the brightness component, and amplifies the high-frequency com component of the brightness component while the low frequency component nent is weakened; where the central control module above- is configured to perform the following homomorphic filter processing got brightness component to combine with the saturation and the tint, and transform it back to the RGB color space end To perform Image Enhancement of the Remote Sensing image; an evaluation module, electrically connected to the central control ring module, where the evaluation module is configured to grayscale distribution of the image in the remote sensing image after to obtain image enhancement, and based on the gray values distribution of the image the information entropy of the remote sen to determine the image after image enhancement; where the evaluation model dule further configured to the gray value of the pixel of to obtain the remote sensing image after image enhancement, and on based on the gray value of the pixel the contrast of the remote to determine the sensing image after image enhancement; where the evaluation tie module is also configured to, based on the information tensile entropy and the contrast, the remote improvement score sensing image after image enhancement to determine, and based on the relationship between the improvement score and the preset improvement tering score to determine whether the remote sensing image represents the image improvement has been completed. FIG.1 FIG.2