A Method and System for Monitoring Mangroves in High-Resolution Remote Sensing Images

Abstract

The present invention provides a method and system for monitoring mangroves in high-resolution remote sensing images. Among them, the method includes: performing image acquisition and processing covering the study area to obtain a fused image; calculating the normalized difference water index and the normalized difference vegetation index based on the fused image to obtain a first calculation result; obtaining a key feature layer according to the first calculation result, resampling the fused image to obtain a first resampled image; dividing and screening sub-regions according to the first resampled image, calculating an automatic threshold according to the division result, and extracting the whole-region water body; obtaining a threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water body regions in the whole-region water body; extracting the intertidal zone according to the water body regions in the whole-region water body; performing masking processing on the fused image to obtain a first masking processing result; and obtaining a mangrove distribution map according to the first masking processing result and the threshold for vegetation extraction.

Description

Technical Field

[0001] The present invention relates to the technical field of space remote sensing detection, and particularly relates to a method and system for monitoring mangroves using high-resolution remote sensing images. Background Art

[0002] Mangroves are unique evergreen broad-leaved forests growing on tropical and subtropical coastal beaches, which have multiple ecosystem service functions. Therefore, mangroves have extremely high detection value. Currently, the remote sensing detection technology for mangroves mainly uses the object-oriented classification method for low and medium-resolution remote sensing images.

[0003] In the process of implementing the inventive technical solution in the embodiments of the present application, it is found that the above technology has at least the following technical problems:

[0004] In the prior art, due to the complex mangrove habitat, remote sensing detection using medium and low-resolution images is not applicable to the monitoring of relatively complex areas and small-scale mangrove areas, and the existing remote sensing monitoring methods for mangroves are not applicable to the use of high-resolution remote sensing images, resulting in the technical problem of relatively low accuracy of mangrove remote sensing monitoring. Summary of the Invention

[0005] The embodiments of the present application provide a method and system for monitoring mangroves using high-resolution remote sensing images, which are used to solve the technical problem of relatively low accuracy of mangrove remote sensing monitoring existing in the prior art that the remote sensing monitoring method for mangroves is not applicable to the use of high-resolution remote sensing images.

[0006] In view of the above problems, the embodiments of the present application provide a method and system for monitoring mangroves using high-resolution remote sensing images.

[0007] In the first aspect of the embodiments of the present application, a method for monitoring mangroves using high-resolution remote sensing images is provided. The method includes: performing image acquisition and processing on a study area to obtain a fused image; calculating a normalized difference water index and a normalized difference vegetation index based on the fused image to obtain a first calculation result; obtaining a key feature layer based on the first calculation result, and performing resampling based on the fused image and the key feature layer to obtain a first resampled image; dividing and screening sub-regions based on the first resampled image, calculating an automatic threshold based on the results of the division and screening of the sub-regions, and extracting the whole-region water body based on the automatic threshold calculation result; obtaining a threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water regions in the whole-region water body; extracting the intertidal zone based on the water body regions in the whole-region water body; performing masking processing on the fused image based on the intertidal zone extraction result and sea surface information to obtain a first masking processing result; and obtaining a mangrove distribution map based on the first masking processing result and the vegetation extraction threshold.

[0008] In the second aspect of the embodiments of the present application, a high-resolution remote sensing image mangrove monitoring system is provided. The system includes: a first acquisition unit configured to collect and process images covering a study area to obtain a fused image; a first processing unit configured to calculate a normalized difference water index and a normalized difference vegetation index based on the fused image to obtain a first calculation result; a second processing unit configured to obtain a key feature layer based on the first calculation result, and perform resampling based on the fused image and the key feature layer to obtain a first resampled image; a third processing unit configured to divide and screen sub-regions based on the first resampled image, calculate an automatic threshold based on the division and screening results of the sub-regions, and extract the whole-region water body based on the automatic threshold calculation result; a second acquisition unit configured to obtain a threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water area in the whole-region water body; a fourth processing unit configured to extract the intertidal zone based on the water body area in the whole-region water body; a third acquisition unit configured to perform masking processing on the fused image based on the intertidal zone extraction result and sea surface information to obtain a first masking processing result; a fifth processing unit configured to obtain a mangrove distribution map based on the first masking processing result and the threshold for vegetation extraction.

[0009] In the third aspect of the embodiments of the present application, a high-resolution remote sensing image mangrove monitoring system is provided, including: a processor coupled to a memory, where the memory is used to store a program, and when the program is executed by the processor, the system is enabled to execute the steps of the method as described in the first aspect.

[0010] In the fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method as described in the first aspect are implemented.

[0011] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0012] The technical solution provided by the embodiment of the present application obtains a high-resolution remote sensing image of the area to be detected, performs multi-spectral and panchromatic data fusion processing on the high-resolution remote sensing image to obtain a fused image after fusion, then calculates the normalized difference water index and the normalized difference vegetation index for the pixels in the fused image to obtain a key feature layer, resamples the fused image based on the key feature layer, divides and screens sub-regions based on the resampled image, calculates an automatic threshold according to the division and screening results, and extracts the entire water area of the area to be detected, then obtains a vegetation extraction threshold, extracts the intertidal zone according to the water area, performs masking processing on the fused image according to the intertidal zone extraction result and sea surface information to obtain a first masking processing result, and then extracts the mangrove area based on the vegetation extraction threshold. The embodiment of the present application is based on the fusion processing of multi-spectral and panchromatic bands of high-resolution images, which can reduce the complexity of regional influence. By resampling the fused image based on the key feature layer and all spectral layers, it is possible to divide the intertidal zone suitable for mangrove production, effectively solving the problems of large amounts of high-resolution image data and reduced computer operation efficiency. By dividing the resampled image into sub-regions and obtaining an automatic threshold based on the automatic threshold determination method of the present application based on image features to distinguish water and non-water regions, it is possible to accurately obtain the intertidal zone area suitable for mangrove production. Finally, the vegetation area is extracted according to the extraction threshold of the vegetation area, and further a mangrove distribution map is obtained. The embodiment of the present application constructs a method for remotely monitoring mangrove areas based on high-resolution remote sensing images, which can accurately extract and monitor mangrove areas with complex growth and small areas. By combining methods such as automatic thresholds and DEM data, it is possible to reduce the image processing cost, improve the processing efficiency, and achieve the technical effect of accurately remotely monitoring mangrove areas.

[0013] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other objects, features and advantages of the present application more obvious and understandable, the following specifically gives the specific embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic flow chart of a method for monitoring mangroves using high-resolution remote sensing images provided by an embodiment of the present application;

[0015] Figure 2 It is a flow block diagram of a method for monitoring mangroves using high-resolution remote sensing images provided by an embodiment of the present application;

[0016] Figure 3 It is a schematic flow chart of information security supervision in a method for monitoring mangroves using high-resolution remote sensing images provided by an embodiment of the present application;

[0017] Figure 4Schematic diagram of sub - area division in a high - resolution remote - sensing image mangrove monitoring method provided by an embodiment of the present application;

[0018] Figure 5 Schematic diagram of the extraction result of all water areas in a high - resolution remote - sensing image mangrove monitoring method provided by an embodiment of the present application;

[0019] Figure 6 Schematic diagram of the optimal sub - area of the vegetation automatic threshold in a high - resolution remote - sensing image mangrove monitoring method provided by an embodiment of the present application

[0020] Figure 7 Schematic diagram of the extraction of the first vegetation area in a high - resolution remote - sensing image mangrove monitoring method provided by an embodiment of the present application;

[0021] Figure 8 Schematic diagram of the extraction of the mangrove area in a high - resolution remote - sensing image mangrove monitoring method provided by an embodiment of the present application;

[0022] Figure 9 Schematic diagram of the structure of a high - resolution remote - sensing image mangrove monitoring system provided by an embodiment of the present application;

[0023] Figure 10 Schematic diagram of the structure of an exemplary electronic device in an embodiment of the present application.

[0024] Explanation of reference numerals: The first acquisition unit 11, the first processing unit 12, the second processing unit 13, the third processing unit 14, the second acquisition unit 15, the fourth processing unit 16, the third acquisition unit 17, the fifth processing unit 18, the electronic device 300, the memory 301, the processor 302, the communication interface 303, the bus architecture 304. Detailed implementation manners

[0025] By providing a high - resolution remote - sensing image mangrove monitoring method and system, the embodiment of the present application is used to solve the technical problem that the remote - sensing monitoring method of mangroves in the prior art is not applicable to high - resolution remote - sensing images, resulting in low accuracy of mangrove remote - sensing monitoring.

[0026] Application overview

[0027] Mangroves are unique evergreen broad-leaved forests growing on tropical and subtropical coastal tidal flats. They possess the ecological system characteristics of both the ocean and land, and have various ecosystem service functions, including purifying seawater, preventing erosion, and maintaining biodiversity in estuary areas, thus having extremely high monitoring value. Traditional field investigation methods have problems such as being time-consuming, laborious, and difficult to obtain large observation scales. Remote sensing technology, with its advantages of low acquisition cost, short monitoring cycle, and wide observation range, has become one of the most important means for mangrove monitoring. Therefore, studying the extraction method of mangroves from remote sensing images is a very valuable task.

[0028] Due to advantages such as free acquisition, short revisit cycle, and large coverage area, medium and low-resolution satellite data is the most commonly used data source in current mangrove wetland remote sensing monitoring and has been widely applied in various studies. In China, the continuous and concentrated area of mangroves is small, the distribution is scattered, and the habitat is complex. The average area of mangrove patches in some regions is only 4.1 hm 2 . If the area measurement error is to be kept within 5%, the corresponding image resolution should be better than 5 meters. Therefore, medium and low-resolution images, especially the limited spatial resolution of the Landsat satellite series, are not suitable for mangrove monitoring at the regional and local scales in China.

[0029] Since the swath width of high-spatial-resolution data is small, usually several scenes of data are required to cover the entire area. To ensure the timeliness of monitoring, most will use two or more data sources. Therefore, the image situation in the area is relatively complex and the data volume is large. The extraction of key intermediate land cover types such as water bodies and vegetation usually adopts threshold methods and machine learning methods. The method of manually adjusting the threshold has a large workload, so an automatic threshold method is considered to determine the key threshold. There are more interfering features of water bodies and vegetation in the image. Therefore, it is obviously not advisable to use a global method to find the optimal threshold, and the accuracy of the results obtained is usually unstable, with many false changes or missed extraction patches; limited by the results of the initial threshold, the accuracy of the "global-local" iterative method will also be affected to a certain extent. Therefore, using the global threshold result and historical thematic vectors as a reference to determine the optimal threshold area has problems of unstable accuracy and difficulty in engineering implementation. Current mangrove extraction algorithms mostly target medium and low-resolution remote sensing images, use their rich spectral information, perform various band operations, and obtain a large number of spectral features to distinguish mangroves from interfering land cover types. The number of bands of high-resolution images is limited and the spectral information is weak. However, there are obvious spatial feature differences between mangroves and interfering land cover types. Therefore, the current extraction algorithms dominated by spectral features cannot be fully applied to the extraction of high-resolution mangrove information.

[0030] In the prior art, due to the complex habitat of mangroves, remote sensing detection using medium- and low-resolution images is not applicable to the monitoring of relatively complex areas and small-scale mangrove areas. Moreover, the number of bands of high-resolution images in the prior art is limited and the spectral information is weak. However, there are obvious spatial feature differences between mangroves and disturbed land classes. The extraction algorithm dominated by spectral features for mangroves is not applicable to high-resolution remote sensing images, resulting in the technical problem of low accuracy in mangrove remote sensing monitoring.

[0031] In view of the above technical problems, the general idea of the technical solution provided in this application is as follows:

[0032] Collect and process images covering the study area to obtain a fused image; calculate the normalized difference water index and the normalized difference vegetation index based on the fused image to obtain a first calculation result; obtain a key feature layer based on the first calculation result, and perform resampling based on the fused image and the key feature layer to obtain a first resampled image; divide and screen sub-regions based on the first resampled image, calculate an automatic threshold based on the results of the sub-region division and screening, and extract the whole-region water body based on the automatic threshold calculation result; obtain a threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water areas in the whole-region water body; extract the intertidal zone based on the water body areas in the whole-region water body; perform masking processing on the fused image based on the intertidal zone extraction result and sea surface information to obtain a first masking processing result; and obtain a mangrove distribution map based on the first masking processing result and the vegetation extraction threshold.

[0033] After introducing the basic principle of this application, hereinafter, the technical solutions in the embodiments of this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application. Additionally, it should be noted that for the sake of description, only parts related to this application are shown in the accompanying drawings rather than all of them.

[0034] Embodiment 1

[0035] As Figure 1 shown, the embodiment of this application provides a method for monitoring mangroves using high-resolution remote sensing images, and the method includes:

[0036] S100: Collect and process images covering the study area to obtain a fused image;

[0037] Figure 2FIG. 0 shows a possible flowchart of the method provided by the embodiments of the present application. Specifically, the covered research area is the area where remote sensing monitoring of mangrove areas needs to be carried out, generally coastal and island areas. In the embodiments of the present application, in order to monitor and obtain a mangrove distribution area with higher accuracy, remote sensing monitoring is carried out based on high-resolution remote sensing images. Therefore, high-resolution remote sensing images of the covered research area are acquired. Optimally, high-resolution remote sensing image products that have been preprocessed, radiometrically corrected, and have a cloud cover of less than 10% are acquired to improve the accuracy of remote sensing detection.

[0038] Since the data width of high-resolution remote sensing images is small, several scenes of data are usually required to cover the entire area. In order to keep the remote sensing monitoring images as images of the same time period, generally two or more high-resolution image data sources are acquired, and then multiple high-resolution images of different data sources are subjected to multispectral and panchromatic band fusion processing to obtain a fused image. Exemplarily, the above-mentioned fused image can also be a multispectral image, which can be selected according to the actual remote sensing image and business requirements.

[0039] Further, in order to more accurately distinguish the water bodies in the ocean area and the land area after the water area is acquired, digital elevation model (DEM) data covering the research area is collected at the same time. Preferably, DEM data with a resolution of 30 meters or more is acquired.

[0040] S200: Calculate the normalized difference water index and the normalized difference vegetation index according to the fused image to obtain a first calculation result;

[0041] Specifically, according to the above-mentioned fused image obtained by the fusion process, for each pixel in the fused image, the normalized difference water index (NDWI) and the normalized difference vegetation index (NDVI) are calculated. The calculation is performed by the following formula:

[0042]

[0043]

[0044] where DNgreen, DNred, and DNNIR are the spectral values of the green band, red band, and near-infrared band, respectively. The normalized difference water index and the normalized difference vegetation index obtained based on the above calculations are used as the first calculation result.

[0045] S300: Obtain a key feature layer according to the first calculation result, and perform resampling based on the fused image and the key feature layer to obtain a first resampled image;

[0046] Specifically, obtain a key feature layer based on the normalized water index and the normalized vegetation index calculated in step S200, and then perform resampling on the fused image obtained in step S100 for all bands, including multi-spectral bands and key feature bands, to obtain a first resampled image. The first resampled image is the image near the coastline within the fused image. For all band layers, including multi-spectral bands and key feature bands, it can significantly improve the operation efficiency in the next optimal sub-region and optimal threshold iteration.

[0047] Preferably, resample to obtain an image with a resolution of 5 meters in the multi-spectral image, and use the pixel value closest to the center pixel of the source matrix in the original multi-spectral image as the pixel value of the first resampled image.

[0048] S400: Perform sub-region division and screening according to the first resampled image, perform automatic threshold calculation according to the sub-region division and screening results, and extract the whole-region water body according to the automatic threshold calculation result;

[0049] Step S400 in the method provided by the embodiments of the present application includes:

[0050] S410: Obtain a first predetermined sub-region division rule;

[0051] S420: Perform sub-region division of the first resampled image according to the first predetermined sub-region division rule to obtain a first sub-region division result;

[0052] S430: Determine whether the first sub-region division result meets a first preset condition. When the first sub-region division result meets the first preset condition, obtain valid sub-regions;

[0053] S440: Screen and obtain a first sub-region and a second sub-region according to the valid sub-regions;

[0054] S450: Perform the automatic threshold calculation according to the first sub-region and the second sub-region, and extract the entire water area according to the automatic threshold calculation result.

[0055] Specifically, due to the large amount of data in high-resolution remote sensing image data and the relatively complex image conditions, the method of manually setting thresholds to extract vegetation and water bodies involves a large amount of work and has low universality. Therefore, an automatic threshold method is adopted to determine the key thresholds. However, since there are many interfering features of water bodies and vegetation in the image, it is not advisable to directly adopt a global threshold for the first resampled area, and the accuracy and reliability of the obtained results are relatively low. Therefore, it is necessary to segment the first resampled image, and then set automatic thresholds according to each segmented area, and iteratively search for the global optimal threshold in the automatic thresholds of each segmented area for extraction.

[0056] As Figure 3 shown, according to the first predetermined sub-region division rule, the first resampled image obtained by resampling is divided into sub-regions. The first predetermined sub-region division rule includes dividing the first resampled image into several sub-regions of equal size. Exemplarily, the first resampled image is divided into a checkerboard pattern according to the first predetermined sub-region division rule, and the scale size of the division is 5000*5000 pixel, resulting in a plurality of sub-regions of the same size.

[0057] Further, it is determined whether each of the segmented sub-regions meets the first preset condition, and then it is determined whether the images within each sub-region can be used or are suitable for remote sensing detection. Exemplarily, the first preset condition includes the proportion of effective pixels within the divided sub-region. It is necessary to ensure that the effective pixels are greater than a certain area within a certain sub-region before the sub-region can be used as an effective sub-region for subsequent processing. Preferably, the first preset condition is that a sub-region with an effective pixel area greater than 70% in the divided sub-region is an effective sub-region.

[0058] Step S440 in the method provided by the embodiment of the present application includes:

[0059] S441: Conduct key feature statistics according to the effective sub-regions to obtain a first key feature statistics result, where the first key feature statistics result includes a normalized water index sequence;

[0060] S442: Obtain an optimal sub-region screening rule, and obtain the first sub-region and the second sub-region according to the optimal sub-region screening rule and the first key feature statistics result.

[0061] Specifically, first, calculate the normalized difference water index mean NDWI_MEAN of all valid sub-regions. Then, taking the valid sub-regions as units, sort the pixels in the valid sub-regions in ascending order according to the normalized difference water index mean, and record them at an appropriate step size. Exemplarily, in the embodiments of the present application, considering the relationship between the accuracy and scale of the mangrove area, the step size is set to 5%, and the value at the end of the sorting within the interval is recorded according to this step size, obtaining the normalized difference water index sequence [Q5, Q10, Q15......Q95] in each valid sub-region.

[0062] Then, based on the first key feature statistical results including the normalized water body sequence, screen the valid sub-regions according to the optimal sub-region screening rules.

[0063] Preferably, the optimal sub-region screening rules include: (1) The normalized difference water index sequence of the optimal sub-region satisfies Q5 < NDWI_MEAN and Q95 > NDWI_MEAN; (2) Divide the normalized difference water index sequence of each valid sub-region into 4 segments, statistically calculate the difference between the middle segment, i.e., Q75 (75%) and Q25 (25%) in the valid sub-region, and then sort the differences of this middle segment, and select the largest region as the candidate region. It can be considered that the normalized difference water index of the pixels in the candidate region varies greatly, and there are differences between land and sea. Then calculate the slope of the change of the values in the normalized difference water index sequence in the candidate region. If the maximum slope appears in the interval [Q40, Q45, Q50, Q55, Q60], then select this candidate region as the optimal sub-region. (3) Repeat the above optimal sub-region screening steps (1) and (2) to select two optimal sub-regions. Figure 4 Shows a possible screening result of the optimal sub-region of the water body automatic threshold. Figure 4 Among them, the two optimal sub-regions obtained by the water body automatic threshold are within the white square frames.

[0064] Within the optimal sub-region screening rules, in the valid sub-regions that do not meet step (1), they may be areas with extremely little water such as all land or areas with all water such as all sea, without coastal areas, so there is no possibility of detecting mangrove areas, thus screening is carried out. Step (2) can screen out sub-regions with a large change in the normalized difference water index. In other words, it can select sub-regions that simultaneously have areas with extremely little water and all water areas such as land and sea, and further can screen out sub-regions with a uniform proportion of effective land and sea areas as the optimal sub-regions.

[0065] The embodiments of the present application can determine the key thresholds for each divided region based on the automatic threshold by dividing the resampling impact, improving the processing efficiency. Then, based on the optimal sub-region screening rules, the optimal sub-regions are screened, and it can screen out valid sub-regions with a uniform proportion of effective land and sea areas as the optimal sub-regions, further accurately extract the sub-regions where mangrove areas exist, improving the accuracy of remote sensing monitoring while reducing the processing cost and improving the processing efficiency.

[0066] Step S450 in the method provided by the embodiment of the present application includes:

[0067] S451: Obtain a water body extraction threshold according to the automatic threshold calculation result;

[0068] S452: Determine whether the normalized water body index of the pixel in the first resampled image is greater than the water body extraction threshold;

[0069] S453: When the normalized water body index of the pixel in the first resampled image is greater than the water body extraction threshold, divide the area corresponding to the first resampled image into a water body area;

[0070] S454: When the normalized water body index of the pixel in the first resampled image is less than or equal to the water body extraction threshold, divide the area corresponding to the first resampled image into a non-water body area;

[0071] S455: Extract all water areas according to the division results of the water body area and the non-water body area.

[0072] Specifically, after obtaining the optimal first sub-region and second sub-region, perform multi-scale segmentation on the first sub-region and the second sub-region, and use an automatic threshold calculation method to obtain a water body extraction threshold T water , and divide the water body area of the first resampled image obtained in step S300. Among them, if the normalized water body index of the pixel in the first resampled image is greater than the water body extraction threshold T water , it is divided into a water body area. If the normalized water body index of the pixel is less than the water body extraction threshold T water , it is divided into a non-water body area. Finally, extract all water areas in the first resampled image according to the division results of the water body area and the non-water body area. Figure 5 Shows a possible result of extracting all water areas, Figure 5 in which a water body area and a land area can be obtained.

[0073] Exemplarily, in the embodiment of the present application, the method for multi-scale segmentation of the first sub-region and the second sub-region is based on the region merging method with the smallest regional heterogeneity, gradually merging from the initial single pixel into smaller image objects, and then gradually merging into larger image objects to complete the final segmentation.

[0074] Further, in the embodiment of the present application, the K-means clustering algorithm is used to obtain the optimal water body threshold T water, but not limited to this. The embodiments of the present application adopt a method for automatically determining key thresholds based on the image features themselves. First, an optimal sub-region suitable for the automatic threshold algorithm is constructed to obtain the automatic threshold, and further, a globally optimal threshold can be obtained, achieving the technical effect of improving the extraction accuracy of water body regions and reducing the influence of pseudo-changes and other factors.

[0075] S500: Obtain the threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water body regions in the global water body;

[0076] Specifically, based on the non-water body regions obtained in step S400, the method for calculating the water body extraction threshold is used to calculate and prepare the extraction threshold, as follows:

[0077] First, based on the above non-water body regions, partition them to obtain multiple sub-regions. Then calculate the mean normalized difference vegetation index NDVI_MEAN of all sub-regions. Then, taking the sub-regions as units, sort the pixels within each sub-region in ascending order according to the mean normalized difference vegetation index of the pixels, and record them at an appropriate step size. Exemplarily, set the step size to 5%, and record the value at the end of the sorting within the recording interval according to this step size to obtain the normalized difference vegetation index sequence [P5, P10, P15......P95] within each sub-region.

[0078] Then, based on the normalized difference vegetation index sequence, screen the sub-regions according to the optimal sub-region screening rules.

[0079] Similar to the above content, the optimal sub-region screening rules include: (1) The normalized difference vegetation index sequence of the optimal sub-region satisfies P5 < NDVI_MEAN and P95 > NDVI_MEAN, excluding sub-regions that are all vegetation regions or non-vegetation regions; (2) Divide the normalized difference vegetation index sequence of each sub-region into 4 segments, statistically calculate the difference between the middle segment P75 (75%) and P25 (25%) within the sub-region, and then sort the differences of this middle segment, and select the largest region as the candidate sub-region. It can be considered that the normalized difference vegetation indices of the pixels within the candidate sub-region vary greatly, and there are vegetation-covered regions and non-vegetation-covered regions. Then calculate the slope of the change in the values within the normalized difference vegetation index sequence within the candidate sub-region. If the maximum slope appears in the interval [P40, P45, P50, P55, P60], then select this candidate sub-region as the optimal sub-region. (3) Repeat the above optimal sub-region screening steps (1) and (2) to select two optimal sub-regions. Figure 6 Shows a possible screening result of the optimal sub-region for the automatic vegetation threshold, Figure 6 The white square in is the optimal sub-region obtained based on the automatic vegetation threshold.

[0080] Based on the two optimal sub-regions, perform multi-scale segmentation and use the automatic threshold calculation method to obtain the vegetation extraction threshold Tvegetation 。

[0081] S600: Extract the intertidal zone according to the water body area in the global water body;

[0082] Step S600 in the method provided by the embodiment of the present application further includes:

[0083] S610: Obtain the water body patch area constraint parameter and the DEM value constraint parameter;

[0084] S620: Extract the intertidal zone of the water body area according to the water body patch area constraint parameter and the DEM value constraint parameter.

[0085] Specifically, to accurately distinguish the water body in the ocean area from the water body in the land area, the embodiment of the present application also uses the aforementioned DEM data to extract the intertidal zone area. Specifically, based on the water body area obtained in step S400, the water body patch area is obtained, and the DEM value constraint parameter is obtained based on the DEM data. Exemplarily, the DEM value constraint parameter is that the water body object with a DEM value lower than 10 meters, the largest area and not completely surrounded by non-water bodies in the water body area is used as the sea surface. According to this DEM value constraint parameter, the sea surface range in the water body area image is obtained, and then the non-water body area is segmented at multiple scales, and the minimum distance from the minimum DEM value of the non-water body area pixels to the sea surface area is statistically calculated, and the intertidal zone area adjacent to the sea is extracted. Mangroves usually grow on the coastal intertidal zone. How to determine the range of the intertidal zone through sea surface information is of great significance for the extraction of mangroves.

[0086] S700: Perform masking processing on the fused image according to the intertidal zone extraction result and the sea surface information to obtain the first masking processing result;

[0087] Specifically, the above-mentioned intertidal zone extraction result is used to perform masking processing on the fused image. Exemplarily, during the masking process, the masking pixels in the intertidal zone area are set to 1, and other areas are set to 0, and the area within the mask is obtained as the above-mentioned first masking processing result.

[0088] S800: Obtain the mangrove distribution map according to the first masking processing result and the threshold for vegetation extraction.

[0089] Step S800 in the method provided by the embodiment of the present application includes:

[0090] S810: Extract the vegetation area in the masked area according to the threshold for vegetation extraction to obtain the first vegetation area;

[0091] S820: Calculate the canny edge feature of the first vegetation area;

[0092] S830: Obtain a first predetermined segmentation scale, perform multi-scale segmentation on the first vegetation area according to the first predetermined segmentation scale, and extract mangrove patches based on the canny edge features and hue values;

[0093] S840: Regularize the extracted mangrove patches to obtain the mangrove distribution map.

[0094] Specifically, based on the above first masking processing result, use the vegetation extraction threshold T obtained in step S500 vegetation , perform feature extraction on the masked area, and extract the vegetation area of the intertidal zone as the first vegetation area. Figure 7 Shows a possible extraction result of the first vegetation area, Figure 7 The white area at the connection part of land and water in [Figure] is the intertidal zone area.

[0095] Based on the extracted first vegetation area, calculate the canny edge features of the vegetation area, perform multi-scale segmentation on the vegetation area, extract mangroves using canny feature values and hue values, and regularize the extracted mangrove patches to obtain the final mangrove distribution map. Figure 8 shows a possible extraction result of the mangrove area, Figure 8 The white part in [Figure] is the mangrove area.

[0096] Exemplarily, the above regularization processing includes merging or removing fragmented patches, morphological growth to eliminate deformed patches, etc., to improve the accuracy of the patches in the mangrove distribution area.

[0097] Next, the application of the method provided in this application in an actual scenario will be described to better understand the technical solution of this application, but it is not a limitation of this application.

[0098] Step 1: Image acquisition and processing

[0099] Collect one or more of the GF2 data source, GF1 data source, and ZY 3 data source covering the study area in the coastal area as the preprocessing-level radiometric correction products in the internal data source. At the same time, perform fusion processing on the multi-spectral and panchromatic bands to obtain the fused image data in units of scenes, with a resolution of 1m, 2m, or 2.1m, including the blue band, green band, red band, and near-infrared band. Collect DEM data covering the study area, with a resolution of 12.5 meters, covering the entire study area.

[0100] Step 2: Key feature calculation

[0101] Based on the fused image, calculate the Normalized Difference Water Index (NDWI) and the Normalized Difference Vegetation Index (NDVI).

[0102]

[0103]

[0104] Among them, DNgreen, DNred, and DNNir are the spectral values of the green band, red band, and near-infrared band, respectively.

[0105] Step 3: Image resampling

[0106] Resample the fused image obtained in Step 1 and the key feature layer obtained in Step 2 to 5 meters. Use the pixel value closest to the center pixel of the source matrix as the pixel value of the resampled image.

[0107] Step 4: Calculate the key threshold for water body extraction and extract the whole-region water body.

[0108] Perform checkerboard segmentation on the resampled image, with the scale of the checkerboard segmentation being 5000*5000 pixels. Remove the pixels with a blue band gray value lower than 1 in the sub-region, calculate the proportion of the area of the remaining pixels to the total area of the sub-region. If this proportion is greater than 70%, it is a valid sub-region.

[0109] Statistically calculate the mean NDWI_MEAN of the normalized difference water index of all valid sub-regions. Sort the normalized difference water index of the pixels included in each sub-region from small to large. Record the value at the end of the sorting within the interval at a 5% step size to obtain the normalized difference water index sequence [Q5, Q10, Q15......Q95].

[0110] Exclude the sub-regions in which the normalized index sequence values in the sub-region satisfy Q5>NDWI_MEAN or Q95<NDWI_MEAN. This part of the region is a possible all-water body region or a region with extremely little water body. Sequentially calculate the difference between Q75 and Q25 of the normalized index sequence values of each sub-region and sort them. Select the region with the largest difference as the candidate region. Calculate the slope of the change in the normalized index sequence values within the candidate region. If the maximum slope appears within the interval [Q40, Q45, Q50, Q55, Q60], then this candidate region is the optimal sub-region. Repeat the above optimal sub-region screening steps to select two optimal sub-regions.

[0111] Perform multi-scale segmentation on the selected sub-regions. This algorithm is based on the region merging algorithm with the minimum regional heterogeneity. It gradually merges from the initial single pixel into smaller image objects, and then gradually merges into larger image objects to complete the final segmentation. For example: If s1 and s2 are merged into s, then the formula for the regional heterogeneity of s is as follows:

[0112] f = w color .h color +(1 - w color ).h shape

[0113]

[0114] h shape = w compct .h compct +(1 - w compct ).h smooth

[0115]

[0116]

[0117] where f, w color , w compct are the merging heterogeneity (segmentation scale), spectral weight, and compactness weight respectively, and h color and h shape are the spectral heterogeneity and shape heterogeneity of the merged patches respectively. c is the layer, w c is the weight of layer c, and σ c is the standard deviation of the spectral values of layer c. obj1 and obj2 represent the smaller image objects for merging, and merge represents the larger merged image object. n refers to the object size, l refers to the perimeter of the object, and b refers to the perimeter of the circumscribed rectangle of the object. In this example, the segmentation scale is set to 10, the spectral weight is set to 0.9, and the compactness weight is set to 0.5.

[0118] (1) Statistically calculate the mean of the normalized water body index of the objects in the sub - area, and use the K - means clustering algorithm to obtain the optimal water body threshold T water , and the specific method is as follows:

[0119] (2) Statistically calculate the mean of NDWI of the objects in the optimal sub - area, and use this as the initial classification mean to divide the area into two categories: target and background.

[0120] (3) Calculate the means μ0 and μ1 of the two categories after classification, and use μ0 and μ1 as the two initial clustering centers Z1(I) and Z2(I) of the K - means clustering algorithm.

[0121] Iteratively update the clustering centers using the K - means clustering algorithm.

[0122] (4) If there is no change in the clustering centers of two adjacent iterations, it indicates that the clustering criterion function has converged, and at this time, stop the iterative calculation. Otherwise, return to step (3) and continue the calculation.

[0123] (5) The optimal segmentation threshold of the image is: T* = 1 / 2(Z1 + Z2). Where Z1 and Z2 are the final clustering centers of the target class and the foreground class respectively.

[0124] Perform water area division on the resampled image obtained in step 3. If the normalized difference water index of a pixel is greater than T water , it is divided into the water area; otherwise, it is a non-water area.

[0125] Step 5: Calculate the key threshold for vegetation extraction.

[0126] For the non-water area obtained in step 4, perform zoning, and screen the optimal sub-areas according to the rules in step 4. The key feature used is the normalized difference vegetation index (NDVI) to obtain the threshold T for vegetation extraction vegetation .

[0127] Step 6: Extract the intertidal zone suitable for mangrove growth

[0128] Based on the water area determined in step 4, assign the water body object with a DEM value lower than 10 meters, the largest area and not completely surrounded by non-water areas to the sea surface. Perform multi-scale segmentation on the non-water area, with a segmentation scale of 100, a spectral weight of 0.9, and a compactness of 0.5. Statistically calculate the minimum DEM value min(dem) of the pixels within the object and the closest distance min(distance_sea) to the sea surface, and assign the object that satisfies min(dem) < 10 and min(distance_sea) < 5000m to the intertidal zone.

[0129] Step 7: Extract intertidal zone vegetation.

[0130] Based on the mangrove intertidal zone and sea surface information obtained in step 6, perform masking processing on the image obtained in step 1. For the area within the mask, use the threshold segmentation algorithm and extract the vegetation distribution area according to the vegetation extraction threshold T obtained in step 5 vegetation .

[0131] Step 8: Extract mangroves.

[0132] Calculate the canny edge features of the vegetation area, perform multi-scale segmentation on the vegetation area, with a segmentation scale of 150, a shape index of 0.1, and a compactness of 0.5. Statistically calculate the mean value of the canny (blue band) edge features and the mean value of the hue of HSI (color conversion of red, green, and blue bands) of all vegetation objects. Assign the remaining objects with both canny feature values and hue values lower than the mean to mangroves, and the remaining objects are other vegetation.

[0133] In this embodiment, in order to ensure the integrity of the extracted patches, a series of post-processing operations are performed on the mangrove patches, including merging or removing fragmented patches, morphological growth to optimize the roundness of the patch boundaries, enhancing the aesthetics of the patches, etc., to obtain the mangrove distribution map. At present, it has been confirmed that the extraction accuracy is above 80%.

[0134] In summary, the embodiments of the present application perform fusion processing on the multi - spectral and panchromatic bands of high - resolution images, enabling the images to retain the optimal resolution while preserving all spectral information. Then, based on all the image band layers and key feature layers, resampling is performed on the fused image, which can quickly obtain the optimal sub - regions, effectively solving the problem that high - resolution images have a large amount of data and reducing the computer operation efficiency. By dividing the resampled image into sub - regions and obtaining the automatic threshold based on the image features using the automatic threshold determination method of the present application, the water body and non - water body regions are distinguished, and the intertidal zone area suitable for mangrove growth can be accurately obtained. Finally, the vegetation area is extracted according to the extraction threshold of the vegetation area, and further the mangrove distribution map is obtained. The embodiments of the present application construct a method for remotely monitoring mangrove areas based on high - resolution remote sensing images, which can accurately extract and monitor mangrove areas with complex growth and distributed in coastal areas. Moreover, by combining methods such as automatic threshold and DEM data, the image processing cost can be reduced, the processing efficiency can be improved, and the technical effect of accurately remotely monitoring mangrove areas can be achieved.

[0135] Embodiment 2

[0136] Based on the same inventive concept as a method for monitoring mangroves in high - resolution remote sensing images in the foregoing embodiment, as Figure 9 shown, the embodiments of the present application provide a system for monitoring mangroves in high - resolution remote sensing images, wherein the system includes:

[0137] A first acquisition unit 11, which is used to perform image acquisition and processing covering the research area to obtain a fused image;

[0138] A first processing unit 12, which is used to calculate the normalized difference water index and the normalized difference vegetation index according to the fused image to obtain a first calculation result;

[0139] A second processing unit 13, which is used to obtain a key feature layer according to the first calculation result, and perform resampling based on the fused image and the key feature layer to obtain a first resampled image;

[0140] A third processing unit 14, which is used to divide and screen sub - regions according to the first resampled image, calculate an automatic threshold according to the results of the sub - region division and screening, and extract the whole - domain water body according to the automatic threshold calculation result;

[0141] A second acquisition unit 15, which is used to obtain the threshold for vegetation extraction according to the first resampled image, the normalized difference vegetation index, and the non - water body regions in the whole - domain water body;

[0142] The fourth processing unit 16 is configured to extract the intertidal zone according to the water area in the global water body;

[0143] The third obtaining unit 17 is configured to perform masking processing on the fused image based on the intertidal zone extraction result and the sea surface information to obtain a first masking processing result;

[0144] The fifth processing unit 18 is configured to obtain a mangrove distribution map according to the first masking processing result and the threshold for vegetation extraction.

[0145] Further, the system further includes:

[0146] The sixth processing unit is configured to extract the vegetation area in the masked area according to the threshold for vegetation extraction to obtain a first vegetation area;

[0147] The seventh processing unit is configured to calculate the canny edge features of the first vegetation area;

[0148] The eighth processing unit is configured to obtain a first predetermined segmentation scale, perform multi-scale segmentation on the first vegetation area according to the first predetermined segmentation scale, and extract mangrove patches according to the canny edge features and the hue values;

[0149] The ninth processing unit is configured to perform regularization processing on the extracted mangrove patches to obtain the mangrove distribution map.

[0150] Further, the system further includes:

[0151] The fourth obtaining unit is configured to obtain a first predetermined sub-region division rule;

[0152] The tenth processing unit is configured to divide the first resampled image into sub-regions according to the first predetermined sub-region division rule to obtain a first sub-region division result;

[0153] The first determination unit is configured to determine whether the first sub-region division result meets a first preset condition. When the first sub-region division result meets the first preset condition, an effective sub-region is obtained;

[0154] The eleventh processing unit is configured to screen and obtain a first sub-region and a second sub-region according to the effective sub-region;

[0155] The twelfth processing unit is configured to perform the automatic threshold calculation according to the first sub-region and the second sub-region, and extract the entire water area according to the automatic threshold calculation result.

[0156] Further, the system further includes:

[0157] A thirteenth processing unit configured to perform key feature statistics based on the valid sub-region to obtain a first key feature statistics result, where the first key feature statistics result includes a normalized water body index sequence;

[0158] A fourteenth processing unit configured to obtain an optimal sub-region screening rule, and obtain the first sub-region and the second sub-region according to the optimal sub-region screening rule and the first key feature statistics result.

[0159] Further, the system further includes:

[0160] A fifteenth processing unit configured to obtain a water body extraction threshold according to the automatic threshold calculation result;

[0161] A second judgment unit configured to judge whether the normalized water body index of the pixel of the first resampled image is greater than the water body extraction threshold;

[0162] A sixteenth processing unit configured to, when the normalized water body index of the pixel of the first resampled image is greater than the water body extraction threshold, divide the region corresponding to the first resampled image into a water body region;

[0163] A seventeenth processing unit configured to, when the normalized water body index of the pixel of the first resampled image is less than or equal to the water body extraction threshold, divide the region corresponding to the first resampled image into a non-water body region;

[0164] A fifth obtaining unit configured to extract all water areas according to the division results of the water body region and the non-water body region.

[0165] Further, the system further includes:

[0166] A sixth obtaining unit configured to obtain a water body patch area constraint parameter and a DEM value constraint parameter;

[0167] An eighteenth processing unit configured to perform intertidal zone extraction of the water body region according to the water body patch area constraint parameter and the DEM value constraint parameter.

[0168] Embodiment III

[0169] Based on the same inventive concept as the high-resolution remote sensing image mangrove monitoring method in the foregoing embodiments, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method in Embodiment 1 is implemented.

[0170] Exemplary electronic device

[0171] Reference is made below Figure 10 to describe the electronic device according to the embodiments of the present application.

[0172] Based on the same inventive concept as the high-resolution remote sensing image mangrove monitoring method in the foregoing embodiments, an embodiment of the present application further provides a high-resolution remote sensing image mangrove monitoring system, including: a processor, the processor is coupled to a memory, and the memory is used to store a program. When the program is executed by the processor, the system is caused to execute the steps of the method described in Embodiment 1.

[0173] The electronic device 300 includes: a processor 302, a communication interface 303, and a memory 301. Optionally, the electronic device 300 may further include a bus architecture 304. Among them, the communication interface 303, the processor 302, and the memory 301 may be interconnected through the bus architecture 304; the bus architecture 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 10 only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0174] The processor 302 may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application solution.

[0175] The communication interface 303 uses any device such as a transceiver for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), wired access networks, etc.

[0176] The memory 301 can be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory can exist independently and be connected to the processor through the bus architecture 304. The memory can also be integrated with the processor.

[0177] Among them, the memory 301 is used to store computer-executable instructions for executing the solution of this application, and is controlled by the processor 302 for execution. The processor 302 is used to execute the computer-executable instructions stored in the memory 301, so as to implement a high-resolution remote sensing image mangrove monitoring method provided by the above embodiments of this application.

[0178] Those of ordinary skill in the art can understand that the various digital numbers such as the first and second involved in this application are only for the convenience of description and are not used to limit the scope of the embodiments of this application, nor do they represent the order of precedence. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one" means one or more. At least two means two or more. "At least one", "any one" or their similar expressions refer to any combination of these items, including any combination of single items (pieces) or plural items (pieces). For example, at least one (piece, type) of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0179] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes one or more available media integrated. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)), etc.

[0180] In the embodiments of the present application, the various illustrative logical units and circuits described can be implemented or operate the described functions through a general-purpose processor, a digital signal processor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination of the above designs. The general-purpose processor can be a microprocessor. Optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented through a combination of computing devices, such as a digital signal processor and a microprocessor, multiple microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0181] The steps of the methods or algorithms described in the embodiments of the present application may be directly embedded in hardware, software units executed by a processor, or a combination of both. The software units may be stored in a RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium in the art. Exemplarily, the storage medium may be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Optionally, the storage medium may also be integrated into the processor. The processor and the storage medium may be provided in an ASIC, and the ASIC may be provided in a terminal. Optionally, the processor and the storage medium may also be provided in different components of the terminal. These computer program instructions may also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, thereby providing instructions for implementing the steps specified in Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0182] Although the present application has been described in conjunction with specific features and their embodiments, it is obvious that various modifications and combinations can be made without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for monitoring mangroves in high - resolution remote sensing images, characterized in that, The method includes: Performing image acquisition and processing for the covered study area, performing panchromatic and multispectral image fusion to obtain a fused image; Calculating the normalized difference water index and the normalized difference vegetation index based on the fused image to obtain a first calculation result; Obtaining a key feature layer based on the first calculation result, and performing resampling based on the fused image and the key feature layer to obtain a first resampled image; Dividing and screening sub-areas based on the first resampled image, calculating an automatic threshold based on the sub-area division and screening results, and extracting the whole-region water body based on the automatic threshold calculation result; Obtaining a threshold for vegetation extraction based on the first resampled image, the normalized difference vegetation index, and the non-water area in the whole-region water body; Performing intertidal zone extraction based on the water body area in the whole-region water body; Performing masking processing on the fused image based on the intertidal zone extraction result and sea surface information to obtain a first masking processing result; Obtaining a mangrove distribution map based on the first masking processing result and the vegetation extraction threshold, including: extracting the vegetation area in the masked area based on the vegetation extraction threshold to obtain a first vegetation area; Calculating the canny edge feature of the first vegetation area; Obtaining a first predetermined segmentation scale, performing multi-scale segmentation on the first vegetation area based on the first predetermined segmentation scale, and extracting mangrove patches based on the canny edge feature and hue value; Performing regularization processing on the extracted mangrove patches to obtain the mangrove distribution map.

2. The method according to claim 1, wherein The step of dividing and screening sub-areas based on the first resampled image, calculating an automatic threshold based on the sub-area division and screening results, and extracting the whole-region water body based on the automatic threshold calculation result further includes: Obtaining a first predetermined sub-area division rule; Dividing the first resampled image into sub-areas based on the first predetermined sub-area division rule to obtain a first sub-area division result; Judging whether the first sub-area division result meets a first preset condition, and when the first sub-area division result meets the first preset condition, obtaining valid sub-areas; Screening and obtaining a first sub-area and a second sub-area based on the valid sub-areas; Performing the automatic threshold calculation based on the first sub-area and the second sub-area, and extracting the whole water area based on the automatic threshold calculation result.

3. The method according to claim 2, wherein The step of screening and obtaining the first sub-area and the second sub-area based on the valid sub-areas further includes: Performing key feature statistics based on the valid sub-areas to obtain a first key feature statistics result, where the first key feature statistics result includes a normalized difference water index sequence; Obtaining an optimal sub-area screening rule, and obtaining the first sub-area and the second sub-area based on the optimal sub-area screening rule and the first key feature statistics result.

4. The method according to claim 2, characterized in that, The method further includes: Obtaining a water body extraction threshold based on the automatic threshold calculation result; Judging whether the normalized difference water index of the pixel of the first resampled image is greater than the water body extraction threshold; When the normalized difference water index of the pixel of the first resampled image is greater than the water body extraction threshold, dividing the area corresponding to the first resampled image into a water body area; When the normalized difference water index of the pixels of the first resampled image is less than or equal to the water body extraction threshold, the area corresponding to the first resampled image is divided into non-water body areas; Extract all water areas according to the division results of the water body areas and the non-water body areas.

5. The method according to claim 1, wherein The method further includes: Obtain a water body patch area constraint parameter and a DEM value constraint parameter; Perform intertidal zone extraction of the water body area according to the water body patch area constraint parameter and the DEM value constraint parameter.

6. The method according to claim 1, wherein The method for regularizing the extracted mangrove patches includes merging or removing fragmented patches and morphological growth to eliminate deformed patches.

7. A high-resolution remote sensing image mangrove monitoring system, characterized in that, The system includes: A first acquisition unit, which is used to collect and process images covering the study area to obtain a fused image; A first processing unit, which is used to calculate the normalized difference water index and the normalized difference vegetation index according to the fused image to obtain a first calculation result; A second processing unit, which is used to obtain a key feature layer according to the first calculation result, and perform resampling based on the fused image and the key feature layer to obtain a first resampled image; A third processing unit, which is used to divide and screen sub-regions according to the first resampled image, calculate an automatic threshold according to the division and screening results of the sub-regions, and extract the whole-region water body according to the automatic threshold calculation result; A second acquisition unit, which is used to obtain a vegetation extraction threshold according to the first resampled image, the normalized difference vegetation index, and the non-water body areas in the whole-region water body; A fourth processing unit, which is used to perform intertidal zone extraction according to the water body areas in the whole-region water body; A third acquisition unit, which is used to perform mask processing on the fused image according to the intertidal zone extraction result and sea surface information to obtain a first mask processing result; A fifth processing unit, which is used to obtain a mangrove distribution map according to the first mask processing result and the vegetation extraction threshold, including: A sixth processing unit, which is used to extract the vegetation area in the masked area according to the vegetation extraction threshold to obtain a first vegetation area; A seventh processing unit, which is used to calculate the canny edge feature of the first vegetation area; An eighth processing unit, which is used to obtain a first predetermined segmentation scale, perform multi-scale segmentation on the first vegetation area according to the first predetermined segmentation scale, and extract mangrove patches according to the canny edge feature and the hue value; A ninth processing unit, which is used to regularize the extracted mangrove patches to obtain the mangrove distribution map.

8. A high-resolution remote sensing image mangrove monitoring system, characterized in that, Includes: A processor, the processor is coupled with a memory, the memory is used to store a program, and when the program is executed by the processor, the system is enabled to execute the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, A computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images