An image processing method for flood control monitoring of the river and lake chief system based on artificial intelligence

Abstract

The present invention belongs to the field of computer image processing, and specifically relates to an image processing method for flood control monitoring of the river and lake chief system based on artificial intelligence. Cameras are deployed at key positions of rivers and lakes, and drones are used to collect images. The images are preprocessed by improving the low-light enhancement and noise reduction algorithms to improve the image quality in low-light and bad weather conditions. The superpixel segmentation technology and the image change detection algorithm are used. First, the gradient information of the image is calculated to adaptively adjust the weight coefficients, and then the features are extracted and classified. The changes are judged by comparing the regions at different time points, and the flood risk is predicted by combining time series analysis. Finally, through the big data platform combined with real-time monitoring information, emergency plans are formulated for different risk levels, and decision-making supports such as resource allocation and public warning suggestions are provided. After the flood, evaluation and summary can also be carried out to improve the flood control strategy. This method can effectively solve many problems of existing flood control monitoring, improve the intelligent level and efficiency, and ensure the safety of people's lives and property and the stable development of the region.

Description

Technical Field

[0001] The present invention belongs to the field of computer image processing, and particularly relates to an image processing method for flood control monitoring of river and lake chiefs based on artificial intelligence. Background Art

[0002] In the flood control work of river and lake chiefs, accurately and timely monitoring of flood conditions is crucial for ensuring the safety of people's lives and property. Existing flood control monitoring mainly relies on manual inspections and simple monitoring equipment, and there are many deficiencies. On the one hand, it is difficult for manual inspections to achieve all-weather and all-round real-time monitoring, and it is easy to have inspection blind spots and missed inspections. On the other hand, under harsh weather conditions such as low light, heavy rain, and thick fog, the images collected by conventional monitoring equipment are often blurred and have many noise points, seriously affecting the acquisition of key flood information. Moreover, for a large amount of monitoring image data, existing analysis and processing methods mostly rely on manual experience for visual observation and judgment, with low efficiency and difficult to guarantee accuracy. This manual-based method is difficult to quickly and accurately identify subtle changes in key flood control locations such as rivers, lakes, and dams, and it is impossible to timely detect potential flood risks, resulting in difficulty in making effective flood control deployments in advance. In addition, in the flood control decision-making process, there is a lack of big data support and intelligent analysis means, and it is difficult to quickly generate a scientific and reasonable command plan according to real-time situations. To overcome these problems, there is an urgent need for an intelligent and efficient flood control monitoring image processing method. Summary of the Invention

[0003] In view of the above technical problems in flood control monitoring, the present invention proposes an image processing method for flood control monitoring of river and lake chiefs based on artificial intelligence.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows, including the following steps:

[0005] S1. First, deploy multiple monitoring cameras at key locations of rivers and lakes to collect monitoring images in real time, and use drones to conduct regular cruises to collect high-angle hydrological images;

[0006] S2. Secondly, improve the image quality under low light and harsh weather conditions through image preprocessing using an improved low-light enhancement algorithm and a noise reduction algorithm;

[0007] S3. Then directly process the collected image data, use superpixel segmentation technology and image change detection algorithms to divide the image into different regions, pay special attention to key flood control locations such as rivers, lakes, and dams, and identify changes in key regions to detect possible flood conditions in advance;

[0008] S4. Finally, through the big data platform combined with real-time monitoring information, provide decision-making suggestions to help river and lake chiefs make accurate flood control decisions.

[0009] Preferably, the implementation steps of using superpixel segmentation technology and image change detection algorithm in step S3 to divide the image into different regions and identify the changes in key regions are as follows:

[0010] S31. First, calculate the gradient information of the input image where represent the gradients of the image in the x and y directions. Based on the gradient information, adaptively adjust the weight coefficients of the spatial and color distances, and introduce an adaptive weight parameter λ(x, y), which is adjusted according to the local gradient magnitude: where α is a regulating factor used to control the influence of the gradient on the weight. When the gradient is large, the weight λ(x, y) will decrease, so that the weight of the spatial distance is higher. The improved distance calculation formula is: where d spatial (p, q) is the spatial distance between pixels p and q, and d color (p, q) is their distance in the color space;

[0011] S32. For each superpixel region R i , extract the color, texture, and edge features of the image. Use this feature information to classify each region into different categories. The extraction formula is: Feature(R i ) = [Color(R i ), Texture(R i ), Edge(R i )], where Color(R i ), Texture(R i ), Edge(R i ) are the color feature, texture feature, and edge feature respectively; then based on the extracted features, classify the superpixel regions through a support vector machine or other classification algorithms;

[0012] S33. Finally, by comparing the superpixel regions at each time point t with the regions at the previous time point t - 1 , determine whether the region has changed. The change detection formula is: When the difference value exceeds the threshold of 0.15, it is considered that the region has changed significantly;

[0013] S34. Finally, once the changes in the key regions are detected, further analyze in combination with time series data to generate a change trend graph. By tracking the changes in multiple time periods, the possible future flood conditions or other risk regions can be predicted.

[0014] Preferably, the implementation method of using the improved low-light enhancement algorithm for image preprocessing in step S2 is:

[0015] S211. First, use the light intensity histogram of the image to calculate the global and local light intensity distributions, and set the global brightness gain coefficient G L and the local brightness gain coefficient G R (x, y), and the formula is as follows: where I max is the maximum brightness value of the image, I avg is the global average brightness, I max (x, y) is the maximum brightness of the local area, and I local (x, y) is the average brightness of the local area;

[0016] S212. Use the adaptive color balance algorithm to restore the color of the image and eliminate the color cast phenomenon caused by insufficient light. Utilize the proportional relationship of the RGB channels to adjust the brightness of each channel to make it balanced, and the formula is as follows: where R avg , G avg , B avg are the average values of the RGB channels of the image;

[0017] S213. Finally, combine the local contrast enhancement method and use the Laplace operator to enhance the edges of the image, and the formula is as follows: where is the Laplace operator, α is the coefficient for adjusting contrast enhancement, and I′(x, y) is the enhanced image.

[0018] Preferably, the implementation method of using the improved noise reduction algorithm for image preprocessing in step S2 is as follows:

[0019] S221. First, perform wavelet transform on the image to decompose the image into a low-frequency subband L(x, y) and multiple high-frequency subbands H i (x, y), and the formula is:

[0020] S222. Perform adaptive threshold noise reduction processing on the high-frequency subband H i (x, y), and set the noise reduction threshold T i , and the formula is as follows: where T i is adaptively adjusted according to the high-frequency noise intensity;

[0021] S223. Perform inverse wavelet transform on the processed high-frequency subband H i ′(x, y) and the low-frequency subband L(x, y) to reconstruct the noise-reduced image, and the formula is as follows:

[0022] S224. Finally, to further improve the noise reduction effect, the non-local mean algorithm is used to refine the denoised image and preserve the texture details of the image. The formula is as follows: Where C(x,y) is the normalization constant, h is the smoothing parameter, Ω(x) is the search window, and j is the displacement of the pixel in the search window relative to the central pixel (x,y), which is used to calculate the similarity with the central pixel and determine the final pixel value after denoising accordingly.

[0023] Preferably, providing decision-making suggestions in step S4 includes formulating corresponding emergency plans for different risk levels, including personnel evacuation, material allocation, and equipment preparation, providing resource allocation suggestions such as mobilizing rescue personnel, materials, and mechanical equipment, etc., so as to quickly respond to possible flood events. Public warning suggestions are put forward to guide the river and lake chiefs on how to issue alerts to the public, notify evacuations or other necessary protective measures. After the flood situation ends, it is recommended to conduct follow-up evaluations and summaries to improve future flood control strategies and emergency management measures.

[0024] Compared with the prior art, the advantages and positive effects of the present invention are as follows. First, the present invention realizes all-round and multi-angle image acquisition by combining the deployment of monitoring cameras and drone cruising at key positions of rivers and lakes, and obtains more comprehensive information. Second, by using the improved low-light enhancement and noise reduction algorithms, the image quality in low light and bad weather is significantly improved, making the key information clearer. Third, the superpixel segmentation technology and the image change detection algorithm can accurately identify the changes in key flood control areas and are applicable to various scenarios. Fourth, the decision-making suggestions provided by the big data platform are scientific and reasonable, effectively guaranteeing the efficiency, accuracy, and timeliness of flood control work, and greatly improving the overall flood control ability. Detailed implementation manners

[0025] In order to more clearly understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described below with reference to the embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0026] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the specific embodiments disclosed in the following specification.

[0027] Example. In past flood control practices, the monitoring mode has faced many severe challenges. The manual inspection method consumes a large amount of manpower and time. Inspectors operate under complex terrain and adverse weather conditions, which not only results in low efficiency but also makes it difficult to conduct non-blind-spot and uninterrupted inspections of the entire basin, often leading to inspection gaps in some hidden areas or during extreme weather. The artificial intelligence-based flood control monitoring image processing method for river and lake chiefs is implemented in this area, hoping to effectively solve many problems existing in the existing flood control monitoring means, improve the intelligent level and efficiency of flood control monitoring in this basin, and effectively guarantee the safety of people's lives and property and the stable development of the regional economy and society.

[0028] First, according to the geographical characteristics of the basin and the distribution of key flood control areas, the deployment locations of monitoring cameras were carefully planned to ensure full coverage of key nodes of the river, weak sections of the dikes, and areas around lakes prone to danger. At the same time, advanced unmanned aerial vehicle (UAV) equipment was equipped to regularly perform cruise missions to obtain high-resolution and multi-angle hydrological images, providing rich data sources for subsequent accurate image analysis.

[0029] An improved low-light enhancement algorithm and noise reduction algorithm are used for image preprocessing to improve the image quality under low light and adverse weather conditions. Through adaptive brightness enhancement and color restoration techniques, combined with local contrast enhancement, the brightness and visible details of the image in low light environments are improved, avoiding overexposure or image distortion caused by traditional enhancement methods. First, using the light intensity histogram of the image, the global and local light intensity distributions are calculated, and the global brightness gain coefficient G L and the local brightness gain coefficient G R (x,y) are set. The formula is as follows: where I max is the maximum brightness value of the image, I avg is the global average brightness, I max (x,y) is the maximum brightness of the local area, and I local (x,y) is the average brightness of the local area; the adaptive color balance algorithm is used to restore the color of the image, eliminating the color cast phenomenon caused by insufficient light. Using the proportional relationship of the RGB channels, the brightness of each channel is adjusted to make it balanced. The formula is as follows: where R avg , G avg , B avg are the average values of the RGB channels of the image; finally, combined with the local contrast enhancement method, the Laplacian operator is used to enhance the edges of the image. The formula is as follows: where is the Laplacian operator, α is the coefficient for adjusting contrast enhancement, and I′(x,y) is the enhanced image.

[0030] Next, a denoising algorithm based on multi-scale analysis is introduced. The image is decomposed by wavelet transform, and the noise is concentrated in the high-frequency part for processing while retaining the image details in the low-frequency part, avoiding the loss of details caused by traditional denoising algorithms. First, the image is subjected to wavelet transform to decompose the image into a low-frequency sub-band L(x, y) and multiple high-frequency sub-bands H i (x, y), and the formula is: For the high-frequency sub-band H i (x, y), adaptive threshold denoising processing is performed, and a denoising threshold T i is set, and the formula is as follows: where T i is adaptively adjusted according to the high-frequency noise intensity; the inverse wavelet transform is performed on the processed high-frequency sub-band H i ′(x, y) and the low-frequency sub-band L(x, y) to reconstruct the denoised image, and the formula is as follows: Finally, in order to further improve the denoising effect, the non-local mean algorithm is used to perform refined processing on the denoised image to preserve the texture details of the image, and the formula is as follows: where C(x, y) is the normalization constant, h is the smoothing parameter, Ω(x) is the search window, j is the displacement of the pixel in the search window relative to the central pixel (x, y), which is used to calculate the similarity with the central pixel and determine the final denoised pixel value accordingly.

[0031] In order to achieve the segmentation of surveillance images and the detection of changes in key areas of rivers and lakes, the present invention locally divides the image through superpixel segmentation technology, uses feature extraction combined with a classifier for region classification, and finally detects region changes through time series analysis. This method avoids using the CNN model and is suitable for applications in scenarios with limited computing resources or insufficient deep learning training data. First, the gradient information of the input image is calculated where represents the gradients of the image in the x and y directions. The weight coefficients of the spatial and color distances are adaptively adjusted based on the gradient information, and an adaptive weight parameter λ(x, y) is introduced, which is adjusted according to the local gradient magnitude: where α is a regulation factor used to control the influence of the gradient on the weight. When the gradient is large, the weight λ(x, y) will decrease, so that the weight of the spatial distance is higher. The improved distance calculation formula is: where d spatial (p, q) is the spatial distance between pixels p and q, and d color (p, q) is their distance in the color space; for each superpixel region R i , the color, texture, and edge features of the image are extracted, and each region is classified into different categories using these feature information. The extraction formula is: Feature(R i ) = [Color(Ri ), Texture(R i ), Edge(R i ), where Color(R i ), Texture(R i ), Edge(R i ) are the color feature, texture feature, and edge feature respectively; then, based on the extracted features, the superpixel regions are classified by a support vector machine or other classification algorithms; finally, by comparing the superpixel regions at each time point t with the regions at the previous time point t - 1, it is determined whether the region has changed. The change detection formula is: When the difference value exceeds the threshold of 0.15, it is considered that the region has changed significantly; finally, once a change in the key region is detected, further analysis is performed in combination with time series data to generate a change trend graph. By tracking the changes over multiple time periods, possible future flood conditions or other risk regions can be predicted.

[0032] Finally, decision-making suggestions are provided by combining real-time monitoring information through a big data platform. For different risk levels, corresponding emergency plans are formulated, including personnel evacuation, material allocation, and equipment preparation, and resource allocation suggestions are provided, such as mobilizing rescue personnel, materials, and mechanical equipment, etc., in order to quickly respond to possible flood events. Public warning suggestions are put forward to guide river and lake chiefs on how to issue alerts to the public, notify evacuations, or other necessary protective measures. After the flood situation ends, it is recommended to conduct follow-up evaluations and summaries to improve future flood control strategies and emergency management measures.

[0033] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical content of the technical solution of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An image processing method for flood control monitoring of river and lake chief system based on artificial intelligence, characterized in that, It includes the following steps: S1. First, deploy multiple monitoring cameras at key positions in rivers and lakes to collect monitoring images in real time, and use drones for regular cruises to collect high-angle hydrological images; S2. Secondly, improve the image quality under low light and adverse weather conditions through image preprocessing using improved low-light enhancement algorithms and noise reduction algorithms; S3. Then directly process the collected image data, use superpixel segmentation technology and image change detection algorithms to divide the image into different regions, pay special attention to the key flood control positions of rivers, lakes, and dams, and identify changes in key regions to detect possible flood conditions in advance; S4. Finally, through the big data platform combined with real-time monitoring information, provide decision-making suggestions to help river and lake chiefs make accurate flood control decisions; The implementation steps of using superpixel segmentation technology and image change detection algorithms in step S3 to divide the image into different regions and identify changes in key regions are as follows: S31. First, calculate the gradient information of the input image Among them represents the gradients of the image in the x and y directions. Based on the gradient information, adaptively adjust the weight coefficients of the spatial and color distances, and introduce an adaptive weight parameter λ(x, y), which is adjusted according to the local gradient magnitude: where α is a regulation factor used to control the influence of the gradient on the weight. When the gradient is large, the weight λ(x, y) will decrease, so that the weight of the spatial distance is higher. The improved distance calculation formula is: where d spatial (p, q) is the spatial distance between pixels p and q, and d color (p, q) is their distance in the color space; S32. For each superpixel region R i , extract the color, texture, and edge features of the image. Use this feature information to classify each region into different categories. The extraction formula is: Feature(R i ) = [Color(R i ), Texture(R i ), Edge(R i )], where Color(R i ), Texture(R i ), Edge(R i ) are the color feature, texture feature, and edge feature respectively; Then, based on the extracted features, classify the superpixel regions through support vector machines or other classification algorithms; S33. Finally, for the superpixel regions at each time point t and the regions at the previous time point t - 1 are compared to determine whether the region has changed. The change detection formula is: When the difference value exceeds the threshold of 0.15, it is considered that a significant change has occurred in this region; S34. Finally, once changes in key regions are detected, conduct further analysis in combination with time series data to generate a change trend chart. By tracking changes over multiple time periods, possible future flood conditions or other risk regions can be predicted.

2. The method for processing flood control monitoring images of the river and lake chief system based on artificial intelligence according to claim 1, wherein, The implementation method of using the improved low-light enhancement algorithm for image preprocessing in step S2 is: S211. First, use the light intensity histogram of the image to calculate the global and local light intensity distributions, and set the global brightness gain coefficient G of the image L and the local brightness gain coefficient G R (x, y), and the formula is as follows: where I max is the maximum brightness value of the image, I avg is the global average brightness, I max (x, y) is the maximum brightness of the local area, I local (x, y) is the average brightness of the local area; S212. Use the adaptive color balance algorithm to restore the color of the image, eliminate the color cast phenomenon caused by insufficient light, and utilize the proportional relationship of the RGB channels to adjust the brightness of each channel to make it balanced. The formula is as follows: where R avg , G avg , B avg are the average values of the RGB channels of the image; S213. Finally, in combination with the local contrast enhancement method, the Laplacian operator is used to enhance the edges of the image, and the formula is as follows: Where is the Laplacian operator, α is the coefficient for adjusting contrast enhancement, and I′(x, y) is the enhanced image.

3. The image processing method for flood control monitoring of the river and lake chief system based on artificial intelligence according to claim 1, characterized in that, The implementation method of using the improved noise reduction algorithm for image preprocessing in step S2 is: S221. First, perform wavelet transform on the image to decompose the image into a low-frequency subband L(x, y) and multiple high-frequency subbands H i (x, y). The formula is as follows: where i represents the directions of different high-frequency subbands; S222. For the high-frequency sub-band H i Perform adaptive threshold denoising processing on (x, y), and set the denoising threshold T i , and the formula is as follows: where T i is adaptively adjusted according to the high-frequency noise intensity; S223. Perform inverse wavelet transform on the processed high-frequency sub-band H i ′(x, y) and the low-frequency sub-band L(x, y) to reconstruct the denoised image. The formula is as follows: S224. Finally, in order to further improve the noise reduction effect, the non-local means algorithm is used to refine the denoised image and preserve the texture details of the image. The formula is as follows: Where C(x, y) is the normalization constant, h is the smoothing parameter, Ω(x) is the search window, j is the displacement of the pixel in the search window relative to the central pixel (x, y), which is used to calculate the similarity with the central pixel and determine the final pixel value after denoising accordingly.

4. A method for processing images of flood control monitoring for the river and lake chief system based on artificial intelligence according to claim 1, characterized in that, The decision-making suggestions provided in step S4 include formulating corresponding emergency plans for different risk levels, including personnel evacuation, material allocation, and equipment preparation, providing resource allocation suggestions to quickly respond to possible flood events, putting forward public warning suggestions to guide river and lake chiefs on how to issue alerts to the public, notify evacuations or other necessary protective measures, and conduct follow-up evaluations and summaries after the flood situation ends to improve future flood control strategies and emergency management measures.

Images