Lake change extraction method and device based on remote sensing image and storage medium

Abstract

The application discloses a lake change extraction method and device based on remote sensing images and a storage medium, and comprises the following steps: acquiring remote sensing images of a target region in multiple time phases, performing normalized water body index calculation on each remote sensing image to obtain water body extraction results of each time phase; constructing the maximum submerged range of the water body in the target region according to the water body in multiple time phases; determining a lake object according to the maximum submerged range; converting the remote sensing image of each time phase into a gray-scale image, calculating the gray-scale variance of the gray-scale image; determining a normalized texture index according to the gray-scale variance; identifying and removing building interference in the gray-scale image according to the normalized texture index; restoring the connected region that is mistakenly removed and meets the connectivity retention condition according to the water body connection condition after the removal, to obtain a lake optimization object; determining the single-time-phase lake area and lake change of each time phase according to the boundary of the lake region, and improving the object consistency of the multi-period lake extraction result and the comparability of the multi-year change sequence.

Description

Technical Field

[0001] This invention relates to the fields of remote sensing hydrological and water resources monitoring, water conservancy remote sensing and geographic information processing, and in particular to a method, device and storage medium for extracting lake changes based on remote sensing images. Background Technology

[0002] In the work of river and lake spatial management and lake change surveys, it is necessary to carry out long-term, high-precision dynamic extraction of large-scale lakes.

[0003] Thresholding methods based on optical remote sensing imagery (such as water index thresholding) are the most commonly used technical approach in operational production due to their simple principles, high computational efficiency, and clear physical meaning. However, in real-world applications involving large areas, multiple years, and multiple lakes, existing thresholding methods face a core technical bottleneck: processing single-period images independently cannot guarantee the consistency and stability of extraction results for the same lake at different times. Improving the consistency and stability of lake extraction has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method, apparatus, and storage medium for extracting lake changes based on remote sensing images, in order to solve the problem of the inability to guarantee the consistency and stability of extraction results for the same lake at different times.

[0005] According to one aspect of the present invention, a method for extracting lake changes based on remote sensing imagery is provided, comprising:

[0006] Acquire remote sensing images of the target area at multiple time phases, calculate the normalized water index based on each remote sensing image, and obtain the water extraction results for each time phase;

[0007] Based on the water body extraction results of the multiple time phases, the maximum inundation range of the water body within the target area is constructed; based on the maximum inundation range, lake objects are determined;

[0008] The remote sensing images of each time phase are converted into grayscale images, and the grayscale variance of the grayscale images is calculated; a normalized texture index is determined based on the grayscale variance; and architectural interference in the grayscale images is identified and removed based on the normalized texture index.

[0009] Based on the connectivity of the removed water bodies, the connected regions in the removed pixels that meet the connectivity preservation conditions are restored to obtain the lake optimization objects;

[0010] Lake regions in remote sensing images of each time phase are identified based on the lake optimization object and normalized water index; the single-time-phase lake area of ​​each time phase is determined based on the boundary of the lake region; and lake changes are determined based on the single-time-phase lake area of ​​the multiple time phases.

[0011] According to another aspect of the present invention, a lake change extraction apparatus based on remote sensing imagery is provided, comprising:

[0012] The remote sensing image acquisition module is used to acquire remote sensing images of the target area at multiple time phases, and calculate the normalized water index based on each remote sensing image to obtain the water extraction results for each time phase.

[0013] The maximum inundation range determination module is used to construct the maximum inundation range of water bodies within the target area based on the water body extraction results of the multiple time phases; and to determine lake objects based on the maximum inundation range.

[0014] The building removal module is used to convert remote sensing images of each time phase into grayscale images, calculate the grayscale variance of the grayscale images, determine the normalized texture index based on the grayscale variance, and identify and remove building interference objects in the grayscale images based on the normalized texture index.

[0015] The recovery module is used to recover the connected regions in the removed pixels that meet the connectivity preservation conditions based on the water body connectivity after removal, so as to obtain the lake optimization object;

[0016] The lake region identification module is used to identify lake regions in remote sensing images of each time phase based on the lake optimization object and the normalized water index.

[0017] A lake area monitoring module is used to determine the single-phase lake area for each time period based on the boundary of the lake region; and to determine lake changes based on the single-phase lake areas of the multiple time periods. According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0018] At least one processor; and

[0019] A memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the lake change extraction method based on remote sensing imagery according to any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the lake change extraction method based on remote sensing imagery as described in any embodiment of the present invention.

[0022] The technical solution of this invention involves acquiring remote sensing images of a target area at multiple time phases, calculating a normalized water index based on each remote sensing image to obtain water extraction results for each time phase, constructing the maximum inundation range of water bodies within the target area based on the water extraction results of the multiple time phases, determining lake objects based on the maximum inundation range, converting the remote sensing images of each time phase into grayscale images, calculating the grayscale variance of the grayscale images, determining a normalized texture index based on the grayscale variance, identifying and removing architectural interference objects in the grayscale images based on the normalized texture index, restoring connected regions in the removed pixels that meet the connectivity preservation condition based on the water body connectivity after removal to obtain optimized lake objects, identifying lake regions in the remote sensing images of each time phase based on the optimized lake objects and the normalized water index, determining the single-time-phase lake area of ​​each time phase based on the boundaries of the lake regions, and determining lake changes based on the single-time-phase lake areas of the multiple time phases. Compared to current methods for lake identification using independent remote sensing images from multiple time periods, the technical solution provided in this invention can calculate the maximum inundation range based on remote sensing images from multiple time periods. This allows for the comprehensive determination of the maximum inundation range for each lake by integrating remote sensing images from multiple time periods. Furthermore, within the maximum inundation range, by calculating the gray-level variance and combining it with a normalized texture index, it enables the removal of obstructions such as buildings in the lake and the restoration of pixels after the removal of obstructions. This ensures that the lake optimization object accurately represents the true water area of ​​the lake. Based on this, the lake area is calculated across multiple time periods to determine lake changes, improving the consistency, temporal comparability, and stability of multi-year change sequences in lake identification.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a method for extracting lake changes based on remote sensing images, provided as an embodiment of the present invention;

[0026] Figure 2 This is a schematic diagram of a remote sensing image processing process provided in an embodiment of the present invention;

[0027] Figure 3A schematic diagram of a lake change extraction device based on remote sensing imagery provided in an embodiment of the present invention;

[0028] Figure 4 A schematic diagram of the structure of an electronic device for implementing the lake change extraction method based on remote sensing images according to an embodiment of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0031] Figure 1 This is a flowchart illustrating a method for extracting lake changes based on remote sensing imagery, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a remote sensing image processing procedure provided by an embodiment of the present invention. This embodiment is applicable to the identification and area change detection of lakes in remote sensing images from multiple time periods. The method can be executed by a lake change extraction device based on remote sensing imagery. This device can be implemented in hardware and / or software and can be configured in electronic devices such as personal computers, laptops, smart devices, mobile devices, or servers. Figure 1 As shown, the method includes:

[0032] Step S101: Acquire remote sensing images of the target area at multiple time phases, calculate the normalized water index based on each remote sensing image, and obtain the water extraction results for each time phase.

[0033] Acquire remote sensing images of the target area across multiple years and time phases, including at least red, green, and near-infrared bands. The target area can be a country or province, etc. The remote sensing images can be satellite-captured, and the scope of the target area is not limited. The time interval can be set as needed, such as daily, monthly, quarterly, or annually. Calculate the Normalized Difference Water Index (NDWI) for each period of remote sensing imagery to obtain multi-period water quality extraction results.

[0034] For the Location in the image The normalized water index of a pixel is defined as:

[0035]

[0036] in, For the first Periodic images in pixels The green light band value at that location, For the first Periodic images in pixels The near-infrared band value at a given location. Assuming there are N time phases in total, then period t represents any one of the N time phases, where t is greater than zero and less than N.

[0037] After obtaining the normalized water index calculation results, apply them according to the threshold. (Usually set to 0) is used for binarization to obtain the binary results of the candidate water bodies. The binarization method typically uses a fixed threshold method.

[0038] ;

[0039] .

[0040] The above formula can generate candidate water body extraction results for each time phase, without requiring the results of a single period to be completely stable at the object level.

[0041] Step S102: Construct the maximum inundation range of the water body within the target area based on the water body extraction results of the multiple time phases; determine the lake object based on the maximum inundation range.

[0042] Based on the water extraction results of multiple time periods, the occurrence of water in each pixel in all periods is statistically analyzed at the pixel scale to construct the maximum inundation range of the target lake, which is any lake in the remote sensing image.

[0043] Assume the total number of periods is Then like a pixel The frequency of water body occurrence is defined as:

[0044] .

[0045] Based on frequency statistics, the binary result of the maximum inundation range can be constructed using any of the following methods. .

[0046] Optionally, the maximum inundation range of the water body within the target area can be constructed based on the water body extraction results of the multiple time phases, and can be calculated using any one of the following three methods:

[0047] (1) Union of water extraction results for each period: Specifically, for each pixel in the target area, calculate the union of water extraction results for multiple time periods to obtain the maximum inundation range in the area (target area).

[0048] The calculation formula is as follows: ;

[0049] .

[0050] A pixel is considered submerged if it has been exposed to water at least once. This process continues, and the union of all pixels in the region that have been exposed to water is taken as the maximum masking range.

[0051] (2) The pixel union that satisfies the minimum occurrence period condition is as follows: for each pixel in the target area, if the number of times it appears in the water extraction results of multiple time phases is greater than the preset number, then the pixel is retained to obtain the maximum flooding range in the area (target area).

[0052] The calculation formula is as follows: ;

[0053] .

[0054] Among them, the preset number of times This indicates the preset minimum number of occurrences. A pixel is considered a water body pixel if it appears at least K times. This process is repeated to obtain the maximum inundation range of water bodies within the region.

[0055] (3) The pixel union that satisfies the minimum occurrence period ratio condition is as follows: for each pixel in the target area, if the ratio of the number of times it appears in the water extraction results of multiple time phases to the total number of time phases is greater than the preset ratio, then the pixel is retained to obtain the maximum flooding range in the area (target area).

[0056] The calculation formula is as follows: ;

[0057] .

[0058] Among them, the preset ratio This represents the preset minimum occurrence period ratio threshold.

[0059] If the number of times water appears in a pixel If the ratio of the pixel to the total number of time phases N is greater than a preset ratio R, then the pixel is considered to be submerged. This process is repeated to obtain the maximum submerged area of ​​the water body within the region.

[0060] The above-described implementation methods can accurately identify the maximum inundation range of water bodies in each region from remote sensing images of multiple time phases in three different ways, thereby improving the accuracy of maximum inundation range identification.

[0061] The above implementation method can optimize the maximum inundation range and improve its accuracy. Instead of directly using the single-period lake boundary as the basis for multi-year variation analysis, this method first utilizes multi-period results to construct a relatively stable maximum inundation range for subsequent object extraction and object-level constraints.

[0062] Optionally, the lake objects are determined based on the maximum inundation range, including: performing connected component extraction, vectorization, and object numbering on the obtained maximum inundation range to form multiple lake objects, and assigning a unique number to each lake object for subsequent multi-period tracking and area statistics.

[0063] Specifically, for the maximum flooding range, connected objects are extracted according to the 8-neighborhood (up, down, left, right, upper left, upper right, lower left, and lower right, a total of 8 neighborhoods) connectivity rule, resulting in a set of objects:

[0064] .

[0065] in, Indicates the first A lake object, The total number of objects.

[0066] For each object Assign a unique identifier This forms a set of object IDs:

[0067] .

[0068] Furthermore, after constructing the maximum inundation range of the water body within the target area based on the water body extraction results from the multiple time phases, the method further includes:

[0069] The maximum flooding range is optimized, and the optimization process includes a combination of at least one or more of the following processes: small patch removal, hole filling, boundary smoothing, or minimum area screening.

[0070] However, the execution of the above operations is conditional on maintaining the main structure of the lake without damage. All subsequent lake extractions (time phases) are limited to the same set of stable objects, thus transforming the independent processing of each phase into multi-phase tracking processing oriented towards a unified set of objects.

[0071] Step S103: Convert the remote sensing image of each time phase into a grayscale image, calculate the grayscale variance of the grayscale image; determine the normalized texture index based on the grayscale variance; identify and remove architectural interference objects in the grayscale image based on the normalized texture index.

[0072] To identify highly textured and heterogeneous interference areas such as buildings, bridges, roads, and embankments, this invention defines a Normalized Texture Index (NTI). First, a grayscale image is constructed based on the input image:

[0073]

[0074] in, This represents the red color channel value in the RGB color coordinates (x, y). This represents the green color channel value in the RGB color coordinates (x, y). This represents the blue color channel value in the RGB color coordinates (x, y). This represents the grayscale value at coordinates (x, y).

[0075] Calculate the grayscale variance within a local window W. The formula is:

[0076]

[0077] in, Represented by pixels Centered on, window size is The local mean. Representing pixels Centered on, window size is The local square mean.

[0078] Then, based on the global normalization upper and lower limits obtained from the sample statistics. and NTI is defined as:

[0079]

[0080] in, This indicates that the result will be truncated to... Interval.

[0081] For a given building disturbance threshold Define a high-value pixel mask for architectural interference:

[0082]

[0083]

[0084] The purpose of this step is to identify interference areas within the object that may belong to buildings or high-texture non-water bodies, providing a basis for subsequent connectivity maintenance corrections.

[0085] Step S104: Based on the water body connectivity after removal, restore the connected regions in the removed pixels that meet the connectivity preservation conditions to obtain the lake optimization object.

[0086] After removing obstructions, determine whether the actual lake has been incorrectly split. If the removal has incorrectly disrupted the lake's connectivity, restore the connected regions that should be retained based on the relationship between the areas of the connected components of adjacent water bodies after the split and the areas of the removed connected components, in order to maintain the actual lake structure.

[0087] Optionally, based on the connectivity of the removed water bodies, the connected regions in the removed pixels that meet the connectivity preservation conditions are restored to obtain the lake optimization object, which can be implemented as follows:

[0088] (1) Construct a water body mask based on the original water body and the building interference.

[0089] For candidate objects First, construct a water mask for the object after removing architectural interference:

[0090]

[0091] in, For object Water mask within the original object in the current period This is the intermediate result after removing architectural interference.

[0092] (2) Determine the water connectivity component of the removed pixels based on the water mask.

[0093] Connectivity component labeling is performed on Wi' to obtain the component set:

[0094]

[0095] At the same time, high-value masks for buildings Connected components are also marked to obtain the set of disconnected connected components:

[0096]

[0097] (3) Determine the pixels to be restored from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter.

[0098] For each removed connected component Analyze its adjacency relationship with the remaining water body connected components. If It must be adjacent to at least two connected water bodies at the same time, and satisfy the following:

[0099] Then the removed connected component is restored, where, , To and The area of ​​adjacent remaining water bodies connected components. Let the area of ​​the removed connected component be denoted. Maintain the scaling parameter for connectivity.

[0100] (4) Perform water body restoration on the pixels to be restored to obtain the lake optimization object.

[0101] The above implementation method can identify the removed pixels by constructing a water body mask, determine whether the removed pixels are located in the middle or at the edge of the water body by calculating the water body connectivity components, and then determine whether the removed connectivity components may be erroneously cut-off areas caused by bridges, roads, etc.; restore the removed connectivity components that meet the connectivity preservation conditions to obtain the lake optimization object, so that the lake optimization object can more realistically represent the real water body range of the lake and improve the accuracy of lake identification.

[0102] Furthermore, after determining the pixels to be recovered from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter, the method further includes:

[0103] The pixels to be recovered are further discriminated based on the median texture index and the median threshold of the building texture, and the corresponding connected components are further removed when the median texture index is greater than the median threshold of the building texture.

[0104] To prevent high-texture building blocks from being incorrectly recovered, this invention can also perform a secondary determination of the NTI median within connected components. For any connected component to be recovered or retained... Its texture index median Defined as:

[0105]

[0106] When the following conditions are met: At that time, the connected component can continue to be removed, where, This represents the median threshold for building textures.

[0107] The above implementation not only removes high-texture areas but also, while eliminating architectural interference, strives to preserve the main connectivity structure, narrow connecting parts, and effective water surface channels of the real lake from being mistakenly severed. It can achieve more accurate building texture recognition based on the median threshold of building texture, thus improving the accuracy of building identification.

[0108] After determining the lake objects based on the maximum inundation range, the process also includes:

[0109] Calculate the proportion of vegetation pixels within the lake object, and when the proportion of vegetation pixels is greater than the object-level vegetation proportion threshold, remove or mark the lake object as a non-target object.

[0110] The Normalized Difference Vegetation Index (NDVI) is calculated for each image period or representative image period, and is defined as follows:

[0111]

[0112] in, For the first Periodic images in pixels Reflectivity in the red light band at that location.

[0113] Given vegetation threshold This allows us to construct binary vegetation results:

[0114]

[0115]

[0116] For candidate objects Calculate the percentage of internal vegetation pixels:

[0117]

[0118] in, Representation Object The total number of pixels contained.

[0119] When object satisfy: Then the object will be removed or marked as a non-target object. The threshold for the proportion of vegetation at the object level.

[0120] The above implementation method can identify vegetation in lake objects by calculating the normalized vegetation index, avoid identifying vegetation as water, and improve the accuracy of lake identification.

[0121] Step S105: Identify lake regions in remote sensing images of each time phase based on the lake optimization object and normalized water index; determine the single-time phase lake area of ​​each time phase based on the boundary of the lake region; determine lake changes based on the single-time phase lake areas of the multiple time phases.

[0122] Optionally, identifying lake regions in remote sensing images for each time phase based on the lake optimization target and normalized water index can be implemented as follows:

[0123] In the remote sensing images of the lake optimization object at each time phase, the normalized water index of each pixel is calculated. The water body inside the lake object is segmented according to the preset water index threshold or the adaptive water index threshold to obtain the boundary of the lake area.

[0124] Within the scope of lake objects with unified numbering, the water bodies inside each object are segmented for each image period, and NDWI (using a preset threshold method or an adaptive threshold method) is used to obtain the lake boundaries and areas for each period, and a multi-year variation sequence is formed based on the object number.

[0125] Within the scope of lake objects with unified numbering, the water bodies inside each object are segmented for each period of imagery to obtain the lake boundaries and areas for each period, and a multi-year variation sequence is formed based on the object number.

[0126] The result of the internal water body of object Oi in the image of period t is defined as:

[0127] ;

[0128] .

[0129] in, For object In the Object-level segmentation threshold in periodic images.

[0130] In a preferred embodiment A fixed threshold can be used; in another preferred embodiment, Otsu's adaptive threshold can be used, and its objective function can be expressed as:

[0131]

[0132] in, and They represent the thresholds respectively. The probability of the next two classes of pixels, and These represent the mean values ​​of the two types of pixels, respectively.

[0133] After obtaining the object No. After the water body is covered, the corresponding lake area can be calculated:

[0134]

[0135] in, This represents the ground area corresponding to a single pixel.

[0136] This can form an object. Multi-year area change series:

[0137]

[0138] Since all periods are extracted from the same numbered object Oi, the area sequence has a clear object correspondence and high temporal comparability.

[0139] The above implementation can accurately segment the actual lake area boundary from the maximum mask range by using a preset water index threshold or an adaptive water index threshold, thereby improving the accuracy of lake water body identification.

[0140] Furthermore, by combining publicly available river vector data and publicly available reservoir vector data, non-lake objects can be removed from lake objects to reduce the impact of rivers and reservoirs on the lake extraction results.

[0141] In a preferred embodiment, it can be determined whether a candidate object significantly intersects with a publicly disclosed river object or a publicly disclosed reservoir object based on the spatial overlay relationship. If there is a significant intersection, it is marked as a non-lake object and removed. This step is a preferred enhancement step and does not constitute a necessary limitation on the object-level constraints and correction principles of this invention.

[0142] The lake change extraction method based on remote sensing imagery provided in this invention involves acquiring remote sensing images of a target area at multiple time phases, calculating a normalized water index for each remote sensing image to obtain water extraction results for each time phase, constructing the maximum inundation range of water bodies within the target area based on the water extraction results of the multiple time phases, determining lake objects based on the maximum inundation range, converting the remote sensing images of each time phase into grayscale images, calculating the grayscale variance of the grayscale images, determining a normalized texture index based on the grayscale variance, identifying and removing building interference objects in the grayscale images based on the normalized texture index, restoring connected regions in the removed pixels that meet the connectivity preservation condition based on the water body connectivity after removal to obtain optimized lake objects, identifying lake regions in the remote sensing images of each time phase based on the optimized lake objects and the normalized water index, determining the single-time-phase lake area of ​​each time phase based on the boundaries of the lake regions, and determining lake changes based on the single-time-phase lake areas of the multiple time phases. Compared to current methods for lake identification using independent remote sensing images from multiple time periods, the technical solution provided in this invention can calculate the maximum inundation range based on remote sensing images from multiple time periods, thus determining the maximum inundation range of each lake by integrating remote sensing images from multiple time periods. Within the maximum inundation range, by calculating the gray-level variance and combining it with the normalized texture index, it is possible to remove building interference objects from the lake and restore the pixels after removing building interference objects, so that the lake optimization object can accurately represent the true water area of ​​the lake. On this basis, the lake area is calculated for multiple time periods to determine lake changes, improving the consistency of the lake object, temporal comparability, and stability of multi-year change sequences.

[0143] By using maximum inundation range as a primary constraint, the consistency of lake extraction results across multiple periods is significantly improved, resolving issues of boundary instability and object mismatch. Unified object numbering enables the stable establishment of multi-year variation sequences for each lake, meeting the operational requirement of year-by-year comparability for each lake. Object-level NDVI and NTI filtering effectively reduces interference from farmland and buildings, and allows for differentiated processing for different lake objects. Connectivity preservation correction prevents real lakes from being mistakenly cut off due to building interference, maintaining the integrity of the lake's physical morphology.

[0144] Figure 3 This is a schematic diagram of a lake change extraction device based on remote sensing imagery provided in an embodiment of the present invention. This embodiment is applicable to situations where lakes in remote sensing images from multiple time periods are identified and their area changes are detected. This lake change extraction device based on remote sensing imagery can be implemented in hardware and / or software, and can be configured in electronic devices such as personal computers, laptops, smart devices, mobile devices, or servers. Figure 3As shown, the device includes: a remote sensing image acquisition module 21, a maximum inundation range determination module 22, a building removal module 23, a restoration module 24, a lake area identification module 25, and a lake area monitoring module 26.

[0145] The remote sensing image acquisition module 21 is used to acquire remote sensing images of the target area at multiple time phases, and calculate the normalized water index based on each remote sensing image to obtain the water extraction result for each time phase.

[0146] The maximum inundation range determination module 22 is used to construct the maximum inundation range of the water body within the target area based on the water body extraction results of the multiple time phases; and to determine the lake object based on the maximum inundation range.

[0147] The building removal module 23 is used to convert remote sensing images of each time phase into grayscale images, calculate the grayscale variance of the grayscale images, determine the normalized texture index based on the grayscale variance, and identify and remove building interference objects in the grayscale images based on the normalized texture index.

[0148] Recovery module 24 is used to recover the connected regions in the removed pixels that meet the connectivity preservation conditions according to the water body connectivity after removal, so as to obtain the lake optimization object;

[0149] The lake region identification module 25 is used to identify lake regions in remote sensing images of each time phase based on the lake optimization object and the normalized water index.

[0150] The lake area monitoring module 26 is used to determine the single-phase lake area of ​​each time period based on the boundary of the lake area; and to determine the lake changes based on the single-phase lake areas of the multiple time periods.

[0151] Based on the above embodiments, optionally, the recovery module 24 is used for:

[0152] Construct a water body mask based on the original water body and the building disturbances;

[0153] The water connectivity component of the removed pixels is determined based on the water mask.

[0154] The pixels to be recovered are determined from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter.

[0155] Water body restoration is performed on the pixels to be restored to obtain the lake optimization object.

[0156] Based on the above embodiments, optionally, a building removal optimization module is also included, which is used to determine the pixels to be restored from the removed pixels according to the water body connectivity component and connectivity retention ratio parameter, and to perform a second discrimination on the pixels to be restored according to the median texture index of the pixels to be restored and the median threshold of building texture, and to continue to remove the corresponding connectivity component when the median texture index is greater than the median threshold of building texture.

[0157] Based on the above embodiments, optionally, a vegetation removal module is also included, which is used to calculate the proportion of vegetation pixels inside the lake object after determining the lake object according to the maximum flooding range, and remove or mark the lake object as a non-target object when the proportion of vegetation pixels is greater than the object-level vegetation proportion threshold.

[0158] Based on the above embodiments, optionally, the lake area identification module 25 is used for:

[0159] In the remote sensing images of the lake optimization object at each time phase, the normalized water index of each pixel is calculated. The water body inside the lake object is segmented according to the preset water index threshold or the adaptive water index threshold to obtain the boundary of the lake area.

[0160] Based on the above embodiments, optionally, the maximum inundation range determination module 22 is used to construct the maximum inundation range of the water body within the target area based on the water body extraction results of the multiple time phases, including:

[0161] For each pixel within the target area, calculate the union of water extraction results from multiple time phases to obtain the maximum inundation extent within the area; or,

[0162] For each pixel within the target area, if it appears more than a preset number of times in the water extraction results across multiple time phases, then that pixel is retained to obtain the maximum inundation range within the area; or...

[0163] For each pixel in the target area, if the ratio of the number of times it appears in the water extraction results of multiple time phases to the total number of time phases is greater than a preset ratio, then the pixel is retained to obtain the maximum flooding range in the area.

[0164] Based on the above embodiments, optionally, a maximum flooding range optimization module is also included, which is used to optimize the maximum flooding range after constructing the maximum flooding range of the water body in the target area based on the water body extraction results of the multiple time phases. The optimization process includes a combination of at least one or more of the following processes: small patch removal, hole filling, boundary smoothing, or minimum area screening.

[0165] The lake change extraction device based on remote sensing imagery provided in this embodiment of the invention includes a remote sensing imagery acquisition module 21, used to acquire remote sensing images of a target area at multiple time phases, calculate a normalized water index based on each remote sensing image, and obtain the water body extraction result for each time phase; a maximum inundation range determination module 22, used to construct the maximum inundation range of the water body within the target area based on the water body extraction results of the multiple time phases; and determine the lake object based on the maximum inundation range; and a building removal module 23, used to convert the remote sensing images of each time phase into grayscale images, calculate the grayscale variance of the grayscale images, and determine the lake object based on the grayscale variance. The system comprises the following modules: a normalized texture index is determined; building interference in the grayscale image is identified and removed based on the normalized texture index; a recovery module 24 is used to recover connected regions in the removed pixels that meet the connectivity preservation condition based on the water connectivity after removal, thus obtaining the lake optimization object; a lake region identification module 25 is used to identify lake regions in the remote sensing image of each time phase based on the lake optimization object and the normalized water index; a lake area monitoring module 26 is used to determine the single-time-phase lake area of ​​each time phase based on the boundary of the lake region; and lake changes are determined based on the single-time-phase lake areas of multiple time phases. Compared with the current lake identification methods based on independent remote sensing images of multiple time phases, the technical solution provided by this embodiment can calculate the maximum inundation range based on remote sensing images of multiple time phases, and realize the determination of the maximum inundation range of each lake by integrating remote sensing images of multiple time phases. Within the maximum inundation range, by calculating the grayscale variance and combining it with the normalized texture index, building interference in the lake is removed, and the pixels after removing building interference are restored, so that the lake optimization object can accurately represent the true water area of ​​the lake. Based on this, the lake area is calculated for multiple time periods to determine lake changes and improve the consistency of lake objects, temporal comparability, and stability of multi-year change sequences.

[0166] The lake change extraction device based on remote sensing imagery provided in this embodiment of the invention can execute the lake change extraction method based on remote sensing imagery provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0167] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0168] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0169] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as a camera, ultrasonic sensor, infrared sensor, etc.; output unit 17, such as various types of speakers, etc.; storage unit 18, such as a disk, solid-state drive, etc.; and communication unit 19, such as a network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0170] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as lake change extraction methods based on remote sensing imagery.

[0171] In some embodiments, the lake change extraction method based on remote sensing imagery can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the lake change extraction method based on remote sensing imagery described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the lake change extraction method based on remote sensing imagery by any other suitable means (e.g., by means of firmware).

[0172] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0173] Computer programs for implementing the lake change extraction method based on remote sensing imagery of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0174] This invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a lake change extraction method based on remote sensing imagery, the method comprising:

[0175] Acquire remote sensing images of the target area at multiple time phases, calculate the normalized water index based on each remote sensing image, and obtain the water extraction results for each time phase;

[0176] Based on the water body extraction results of the multiple time phases, the maximum inundation range of the water body within the target area is constructed; based on the maximum inundation range, lake objects are determined;

[0177] The remote sensing images of each time phase are converted into grayscale images, and the grayscale variance of the grayscale images is calculated; a normalized texture index is determined based on the grayscale variance; and architectural interference in the grayscale images is identified and removed based on the normalized texture index.

[0178] Based on the connectivity of the removed water bodies, the connected regions in the removed pixels that meet the connectivity preservation conditions are restored to obtain the lake optimization objects;

[0179] Lake regions in remote sensing images of each time phase are identified based on the lake optimization object and normalized water index; the single-time-phase lake area of ​​each time phase is determined based on the boundary of the lake region; and lake changes are determined based on the single-time-phase lake area of ​​the multiple time phases.

[0180] Based on the above embodiments, optionally, the connected regions in the removed pixels that meet the connectivity preservation conditions are restored according to the water body connectivity after removal, to obtain the lake optimization object, including:

[0181] Construct a water body mask based on the original water body and the building disturbances;

[0182] The water connectivity component of the removed pixels is determined based on the water mask.

[0183] The pixels to be recovered are determined from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter.

[0184] Water body restoration is performed on the pixels to be restored to obtain the lake optimization object.

[0185] Based on the above embodiments, optionally, after determining the pixels to be restored from the removed pixels according to the water connectivity component and connectivity retention ratio parameter, the method further includes:

[0186] The pixels to be recovered are further discriminated based on the median texture index and the median threshold of the building texture, and the corresponding connected components are further removed when the median texture index is greater than the median threshold of the building texture.

[0187] Based on the above embodiments, optionally, after determining the lake object according to the maximum inundation range, the method further includes:

[0188] Calculate the proportion of vegetation pixels within the lake object, and when the proportion of vegetation pixels is greater than the object-level vegetation proportion threshold, remove or mark the lake object as a non-target object.

[0189] Based on the above embodiments, optionally, identifying lake regions in remote sensing images of each time phase according to the lake optimization object and the normalized water index includes:

[0190] In the remote sensing images of the lake optimization object at each time phase, the normalized water index of each pixel is calculated. The water body inside the lake object is segmented according to the preset water index threshold or the adaptive water index threshold to obtain the boundary of the lake area.

[0191] Based on the above embodiments, optionally, the maximum inundation range of the water body within the target area is constructed according to the water body extraction results of the multiple time phases, including:

[0192] For each pixel within the target area, calculate the union of water extraction results from multiple time phases to obtain the maximum inundation extent within the area; or,

[0193] For each pixel within the target area, if it appears more than a preset number of times in the water extraction results across multiple time phases, then that pixel is retained to obtain the maximum inundation range within the area; or...

[0194] For each pixel in the target area, if the ratio of the number of times it appears in the water extraction results of multiple time phases to the total number of time phases is greater than a preset ratio, then the pixel is retained to obtain the maximum flooding range in the area.

[0195] Based on the above embodiments, optionally, after constructing the maximum inundation range of the water body within the target area according to the water body extraction results of the multiple time phases, the method further includes:

[0196] The maximum flooding range is optimized, and the optimization process includes a combination of at least one or more of the following processes: small patch removal, hole filling, boundary smoothing, or minimum area screening.

[0197] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0198] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0199] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0200] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0201] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0202] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for extracting lake change based on remote sensing image, characterized in that, include: Acquire remote sensing images of the target area at multiple time phases, calculate the normalized water index based on each remote sensing image, and obtain the water extraction results for each time phase; Based on the water body extraction results of the multiple time phases, the maximum inundation range of the water body within the target area is constructed; based on the maximum inundation range, lake objects are determined; The remote sensing images of each time phase are converted into grayscale images, and the grayscale variance of the grayscale images is calculated; a normalized texture index is determined based on the grayscale variance; and architectural interference in the grayscale images is identified and removed based on the normalized texture index. Based on the connectivity of the removed water bodies, the connected regions in the removed pixels that meet the connectivity preservation conditions are restored to obtain the lake optimization objects; Lake regions in remote sensing images of each time phase are identified based on the lake optimization object and normalized water index; the single-time-phase lake area of ​​each time phase is determined based on the boundary of the lake region; and lake changes are determined based on the single-time-phase lake area of ​​the multiple time phases.

2. The method of claim 1, wherein, Based on the connectivity of the removed water bodies, connected regions that meet the connectivity preservation conditions in the removed pixels are restored to obtain lake optimization objects, including: Construct a water body mask based on the original water body and the building disturbances; The water connectivity component of the removed pixels is determined based on the water mask. The pixels to be recovered are determined from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter. Water body restoration is performed on the pixels to be restored to obtain the lake optimization object.

3. The method according to claim 2, characterized in that, After determining the pixels to be recovered from the removed pixels based on the water body connectivity component and connectivity retention ratio parameter, the method further includes: The pixels to be recovered are further discriminated based on the median texture index and the median threshold of the building texture, and the corresponding connected components are further removed when the median texture index is greater than the median threshold of the building texture.

4. The method of claim 1, wherein, After determining the lake objects based on the maximum inundation range, the process also includes: Calculate the proportion of vegetation pixels within the lake object, and when the proportion of vegetation pixels is greater than the object-level vegetation proportion threshold, remove or mark the lake object as a non-target object.

5. The method of claim 1, wherein, Based on the lake optimization target and normalized water index, lake regions in remote sensing images of each time phase are identified, including: In the remote sensing images of the lake optimization object at each time phase, the normalized water index of each pixel is calculated. The water body inside the lake object is segmented according to the preset water index threshold or the adaptive water index threshold to obtain the boundary of the lake area.

6. The method of claim 1, wherein, Based on the water body extraction results from the multiple time phases, the maximum inundation range of the water body within the target area is constructed, including: For each pixel within the target area, calculate the union of water extraction results from multiple time phases to obtain the maximum inundation extent within the area; or, For each pixel within the target area, if it appears more than a preset number of times in the water extraction results across multiple time phases, then that pixel is retained to obtain the maximum inundation range within the area; or... For each pixel in the target area, if the ratio of the number of times it appears in the water extraction results of multiple time phases to the total number of time phases is greater than a preset ratio, then the pixel is retained to obtain the maximum flooding range in the area.

7. The method of claim 6, wherein, After constructing the maximum inundation range of the water body within the target area based on the water body extraction results of the multiple time phases, the method further includes: The maximum flooding range is optimized, and the optimization process includes a combination of at least one or more of the following processes: small patch removal, hole filling, boundary smoothing, or minimum area screening. 8.A device for extracting lake change based on remote sensing image, characterized in that, include: The remote sensing image acquisition module is used to acquire remote sensing images of the target area at multiple time phases, and calculate the normalized water index based on each remote sensing image to obtain the water extraction results for each time phase. The maximum inundation range determination module is used to construct the maximum inundation range of water bodies within the target area based on the water body extraction results of the multiple time phases; and to determine lake objects based on the maximum inundation range. The building removal module is used to convert remote sensing images of each time phase into grayscale images, calculate the grayscale variance of the grayscale images, determine the normalized texture index based on the grayscale variance, and identify and remove building interference objects in the grayscale images based on the normalized texture index. The recovery module is used to recover the connected regions in the removed pixels that meet the connectivity preservation conditions based on the water body connectivity after removal, so as to obtain the lake optimization object; The lake region identification module is used to identify lake regions in remote sensing images of each time phase based on the lake optimization object and the normalized water index. The lake area monitoring module is used to determine the single-phase lake area for each time period based on the boundary of the lake region; and to determine lake changes based on the single-phase lake areas for multiple time periods.

9. An electronic device, comprising: The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, which enables the at least one processor to perform the lake change extraction method based on remote sensing imagery as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the lake change extraction method based on remote sensing imagery as described in any one of claims 1-7.

Images