High-resolution remote sensing image processing method and related equipment

Abstract

The embodiment of the invention provides a high-resolution remote sensing image processing method and related equipment, which can store important edge information of a change area in a remote sensing image and avoid the phenomenon of edge blurring or distortion. The method comprises the following steps: preprocessing a front-time-phase high-resolution remote sensing image and a rear-time-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image; comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result graph; performing connected component analysis on a non-change region corresponding to the initial change detection result graph to obtain a region set; performing morphological reconstruction on the region with the area smaller than a set threshold value in the region set to obtain a hole filling result graph; performing small-area noise removal operation on the hole filling result graph to obtain a small-area noise removal result graph; constructing a reconstruction template covering the target image area based on the initial change detection result graph; and processing the initial change detection result image according to a reconstruction template and the small-area noise removal result image.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image processing, and in particular to a high-resolution remote sensing image processing method and related equipment. Background Art

[0002] With the rapid development of remote sensing technology, high-resolution remote sensing images have been widely used in many fields, including urban planning, disaster assessment, environmental monitoring, and resource management. In these applications, change detection is a crucial technology that helps us understand the changes in surface features over time. However, with the improvement of the resolution of remote sensing images, images contain rich information about objects, such as buildings, roads, vegetation, etc., which may show different shapes or colors at different time points. The data collected by any sensor contains a certain amount of noise from the electronic device itself or the external environment. In high-resolution images, the pixels are small and the noise images are relatively larger. In addition, there are multiple factors such as seasonal changes, weather images, imaging geometry images, and image registration errors. As a result, even if the change detection algorithm performs well in the training, verification, and testing stages, the change information extracted during the project production process still has problems such as false alarms and discontinuous extraction results. These problems not only reduce the accuracy of the change detection results, but also increase the workload of manual review and finishing, reduce the efficiency of production work, and increase costs. Especially in application scenarios that require highly accurate change information extraction, such as land use change detection and vegetation cover change analysis, these problems are particularly obvious.

[0003] In order to meet these challenges, new methods are constantly being explored to improve the efficiency of manual processing of change detection results of high-resolution remote sensing images. Among them, optimization methods based on mathematical morphology have attracted attention due to their powerful denoising capabilities, sensitivity to subtle structures, and simplicity and efficiency. Morphological processing is an image processing method based on set theory. By defining the relationship between structural elements and targets, it effectively removes noise, smoothes edges, etc., thereby optimizing the results of change detection. Among them, the morphological "opening operation" and "closing operation" are widely used tools in remote sensing image change detection, but they still have some limitations. These limitations include the risk of over-processing or under-processing, dependence on preset operation order and parameters, poor processing effect on other noises besides salt and pepper noise, and the processing results may change the edge morphology of the extracted results, further increasing the manual processing work, etc. Summary of the invention

[0004] The embodiment of the present invention provides a high-resolution remote sensing image processing method and related equipment, which controls the interference factors in the remote sensing image through area-constrained hole filling and reconstruction operations after denoising, while avoiding boundary blur or distortion caused by over-expansion and retaining important edge information.

[0005] A first aspect of the present invention provides a method for processing high-resolution remote sensing images, comprising:

[0006] Obtaining the previous phase high-resolution remote sensing image and the next phase high-resolution remote sensing image corresponding to the target image area;

[0007] Preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0008] Comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0009] Performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing area;

[0010] Performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result image, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0011] Performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0012] Constructing a reconstruction template covering the target image area based on the initial change detection result map;

[0013] The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

[0014] A second aspect of the present invention provides a high-resolution remote sensing image processing device, comprising:

[0015] An acquisition module is used to acquire a front-phase high-resolution remote sensing image and a back-phase high-resolution remote sensing image corresponding to a target image area;

[0016] A preprocessing module, used for preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0017] A comparison module, used for comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0018] An analysis module, configured to perform connected component analysis on the non-changing region corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing region;

[0019] A reconstruction module, used for performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result map, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0020] A denoising module, used for performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0021] A construction module, used for constructing a reconstruction template covering the target image area based on the initial change detection result map;

[0022] A processing module is used to process the initial change detection result map according to the reconstruction template and the small area noise removal result map to obtain a target change detection result map.

[0023] In a possible design, the analysis module is specifically used for:

[0024] Extracting a non-changing area from the initial change detection result graph to obtain the non-changing area;

[0025] Determining the connectivity number corresponding to the target image area;

[0026] The non-changing region is processed based on the connectivity number to obtain the region set.

[0027] In a possible design, the reconstruction module is specifically used for:

[0028] Taking each target area in the target area subset as a seed point;

[0029] Performing regional expansion and reconstruction according to the seed point and the boundary of each target region, and determining a set of elements that exceed each target region after expansion and reconstruction;

[0030] adding the set of elements to a marked area;

[0031] Repeat the above steps until the elements in the marked area no longer change, and obtain a reconstruction result image;

[0032] The complement of the reconstruction result map is calculated to obtain the hole filling result map.

[0033] In a possible design, the denoising module is specifically used for:

[0034] Construct a target structure element of a preset size;

[0035] Traversing each pixel point in the hole filling result image based on the target structure element to determine the overlap between the image pixels of the hole filling result image and the structure elements of the target structure element;

[0036] The small area noise removal result map is determined according to the overlapping condition.

[0037] In a possible design, the denoising module determines the small area noise removal result map according to the overlapping condition, including:

[0038] Determine, according to the overlapping condition, target image pixels of the structural elements in the hole filling result image that fail to completely cover the target structural element;

[0039] The target image pixels are deleted from the hole filling result map to obtain the small area noise result map.

[0040] In a possible design, the processing module is specifically used for:

[0041] Determine the small area noise removal result image as a seed, and determine the reconstruction template as a constraint condition;

[0042] Dilation is performed based on the seed and the constraint condition to obtain the target change detection result graph.

[0043] In one possible design, the preprocessing includes at least one of radiation correction, geometric correction, and atmospheric correction.

[0044] A third aspect of an embodiment of the present invention provides an electronic device, including a memory and a processor, wherein the processor is used to implement the steps of the high-resolution remote sensing image processing method as described in the first aspect when executing a computer management program stored in the memory.

[0045] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer management program stored thereon. When the computer management program is executed by a processor, the steps of the high-resolution remote sensing image processing method as described in the first aspect above are implemented.

[0046] In summary, it can be seen that in the embodiment provided by the present invention, the remote sensing image is preprocessed to improve the accuracy and usability of the image. By analyzing the connected components, the hole filling of the area constraint and the reconstruction after denoising, the interference factors in the image are effectively controlled, and the changed area is clearer. At the same time, when expanding, the reconstruction template is used as a constraint to ensure that the expansion reconstruction does not cross the boundary, which can effectively avoid the boundary blur or distortion caused by the expansion phenomenon and retain important edge information. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a process flow of a high-resolution remote sensing image processing method provided by an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of the results obtained by directly applying the change detection algorithm provided by the present invention to a pre-processed remote sensing image;

[0049] Figure 3 A schematic diagram of the final result obtained after processing provided by an embodiment of the present invention;

[0050] Figure 4 A schematic diagram for comparing details provided by an embodiment of the present invention;

[0051] Figure 5 A schematic diagram showing a comparison of two different optimization strategies provided by an embodiment of the present invention - the mainstream opening and closing operation and the method proposed by the present invention - when processing high-resolution remote sensing image change detection results;

[0052] Figure 6 A virtual structural diagram of a high-resolution remote sensing image processing device provided by an embodiment of the present invention;

[0053] Figure 7 A schematic diagram of the hardware structure of a high-resolution remote sensing image processing device provided by an embodiment of the present invention;

[0054] Figure 8 A schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention;

[0055] Fig. 9 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0057] In the following description, specific embodiments of the present invention will be described with reference to steps and symbols performed by one or more computers, unless otherwise stated. Therefore, these steps and operations will be mentioned several times as being performed by a computer, and computer execution referred to herein includes operations by a computer processing unit of electronic signals representing data in a structured form. This operation converts the data or maintains it at a location in the computer's memory system, which can be reconfigured or otherwise change the operation of the computer in a manner familiar to testers in the field. The data structure maintained by the data is a physical location in the memory, which has specific characteristics defined by the data format. However, the principles of the present invention are described in the above text, which does not represent a limitation, and testers in the field will understand that the various steps and operations described below can also be implemented in hardware.

[0058] The principles of the present invention operate using many other general or specific purpose computing, communication environments or configurations. Examples of well-known computing systems, environments and configurations suitable for use in the present invention may include (but are not limited to) handheld phones, personal computers, servers, multi-processor systems, microcomputer-based systems, mainframe computers, and distributed computing environments, including any of the above systems or devices. The terms "first", "second" and "third" in the present invention are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. The following describes a method for processing high-resolution remote sensing images from the perspective of a processing device for high-resolution remote sensing images. The processing device for high-resolution remote sensing images can be a server or a service unit in a server, without specific limitation.

[0059] See also Figure 1 , Figure 1 A schematic flow chart of a method for processing high-resolution remote sensing images provided by an embodiment of the present invention includes:

[0060] 101. Obtain a previous phase high-resolution remote sensing image and a subsequent phase high-resolution remote sensing image corresponding to the target image area.

[0061] In this embodiment, the high-resolution remote sensing image processing device can obtain the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image corresponding to the target image area. Specifically, the high-resolution remote sensing images of the front phase (T1) and the back phase (T2) can be obtained from the satellite data source. These data are the cornerstones of the subsequent processing flow, ensuring that the entire processing process has a reliable data foundation.

[0062] 102. Preprocess the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image.

[0063] In this embodiment, after acquiring the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image, the high-resolution remote sensing image processing device can preprocess the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image respectively to obtain the first remote sensing image and the back-phase high-resolution remote sensing image corresponding to the front-phase high-resolution remote sensing image. The preprocessing includes but is not limited to radiation correction, geometric correction and atmospheric correction, thereby eliminating errors caused by factors such as sensors and atmospheric conditions, thereby improving the accuracy and availability of the image.

[0064] 103. Compare the first remote sensing image with the second remote sensing image to obtain an initial change detection result map.

[0065] In this embodiment, the high-resolution remote sensing image processing device can compare the first remote sensing image with the second remote sensing image to obtain an initial change detection result map. Specifically, a pre-trained change detection model can be used to compare the pre-processed T1 image and T2 image pixel by pixel or region by region, identify the difference area of ​​the two image quality inspections, and generate a preliminary change detection result map. The change detection model extracts image features from the T1 image and the T2 image, for example, by extracting image features from the T1 image and the T2 image through a pre-trained CNN model (such as VGG16, ResNet, etc.), and compares the image features to determine the difference area, thereby generating a preliminary change detection result map.

[0066] 104. Perform connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set.

[0067] In this embodiment, the processing device of the high-resolution remote sensing image can calculate the complement of the initial change detection result map, that is, determine the non-changing area extracted by the change detection model. The purpose of calculating the complement is not only to clarify the unchanged area, but also to enable the subsequent processing to more accurately locate and process the changed area. After that, the complement area is subjected to a fine connected component analysis (the goal of the connected component analysis is to find out the area composed of all connected foregrounds) to identify each independent and mutually disconnected area in the complement, that is, the area set. Specifically, the initial change detection result map can be first subjected to non-changing area extraction, and the number of connections corresponding to the target image area is determined (the number of connections here is obtained according to the definition of connectivity, connectivity refers to whether two or more pixels belong to the same area, here according to the definition of connectivity can be divided into 4 connections or 8 connections, a connected component is the maximum set of all foreground pixels connected to each other in the image, usually corresponding to a single object or feature), and the non-changing area is processed based on the number of connections to obtain the area set.

[0068] 105. Perform morphological reconstruction on each target region in the target region subset to obtain a hole filling result image.

[0069] In this embodiment, after determining the region set, the processing device of the high-resolution remote sensing image can calculate the area of ​​each region in the region set, and select a set of regions with an area less than a set threshold from the region set according to the area, which is defined as a target region subset, and then perform morphological reconstruction on each target region in the target region subset (each target region is a hole and also a connected component) to obtain a reconstructed result image. After obtaining the reconstructed result image, the complement of the reconstructed result image is calculated so that the original hole is filled to obtain an area-constrained hole filling result image. Specifically, each target region in the target region is used as a seed point, and regional expansion and reconstruction is performed according to the seed point and the boundary of each target region, and the element set that exceeds each target region after the expansion and reconstruction is determined, and the element set is added to the marked region, and the execution is repeated until the elements in the marked region no longer change, thereby obtaining the reconstructed result image.

[0070] That is, the processing device of high-resolution remote sensing images can select connected components with an area smaller than a set threshold as seed points. These seed points will be used as starting points in the filling process. For each selected connected component, morphological reconstruction is performed. During the reconstruction process, the seed points are gradually expanded using the dilation operation under the constraint of the hole edge. Each iteration expands the marked area outward by one layer, but only new elements that do not exceed the boundary are added to the marked area. When the marked area no longer changes, it means that a stable state has been reached and the reconstruction is stopped. After the reconstruction is completed, the complement of the reconstruction result is calculated again, so that the original hole is filled and the area-constrained hole filling result is obtained.

[0071] It should be noted that morphological reconstruction gradually extracts and strengthens the structure consistent with the target shape from the labeled image under the constraints of the template image through a carefully designed iterative process. This technology uses geodesic dilation or erosion operations to repeatedly adjust the labeled image until a stable state is reached, so that the final result retains the key features in the template image and highlights the parts of the labeled image that match the target morphology.

[0072] 106. Perform a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image.

[0073] In this embodiment, the processing device of the high-resolution remote sensing image can perform a small area noise removal operation on the hole filling result. Specifically, a target structure element of a preset size can be constructed, such as a structure element of a size of 3*3, and each pixel point in the hole filling result image can be traversed based on the target structure element to determine the overlap between the image pixels of the hole filling result image and the structure elements of the target structure element; and the small area noise removal result image can be determined based on the overlap.

[0074] It should be noted that when determining the small area noise removal result map according to the overlapping situation, the overlapping situation can be first analyzed to determine the target image pixels of the structural elements of the target structural element in the hole filling result map that fail to completely cover the target image pixels of the target object in the target image area, and the target image pixels are deleted from the hole filling result map to obtain the small area noise result. In other words, only when the structural element of the target structural element is completely located inside the target object, the pixel point at the center of the structural element will be retained, otherwise, the pixel point will be regarded as the background or deleted, thereby obtaining the small area noise removal result map.

[0075] 107. Construct a reconstruction template covering the target image area based on the initial change detection result map.

[0076] In this embodiment, the processing device of the high-resolution remote sensing image can construct a reconstruction template covering the entire image area based on the original change detection results. The template includes unchanged areas that need to be retained and changed areas that need to be reconstructed, providing guidance for subsequent morphological reconstruction. The specific method of constructing the reconstruction template is not limited here.

[0077] 108. The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

[0078] In this embodiment, the processing device of the high-resolution remote sensing image can determine the result map of small area noise removal as a seed, and determine the reconstruction template as a constraint; dilation is performed based on the seed and the constraint to obtain the target change detection result map. Specifically, the reconstruction template is used as a constraint condition, the result of small area noise removal is used as a seed, and the geodesic dilation method is used to perform dilation under the constraint condition. The core of this method is that it uses the geodesic distance to measure the shortest path from each pixel to the boundary to ensure that the dilated reconstruction does not cross the boundary.

[0079] In order to describe it more intuitively, the following Figures 2 to 5 To explain:

[0080] See also Figure 2 , Figure 2This is a schematic diagram of the results obtained by directly applying the change detection algorithm provided by the present invention to the preprocessed remote sensing image. At this stage, the change area identified by the change detection algorithm is relatively obvious, but there are still some small areas of noise and unfilled small holes, which affect the accuracy and overall appearance of the results. Figure 3 This is a schematic diagram of the final result obtained after processing according to an embodiment of the present invention. Through the operations of area-constrained hole filling and post-noise reconstruction, the interference factors in the image are effectively controlled and the changed area is clearer.

[0081] Figure 4 This provides a detailed comparison, clearly showing the improvement brought by post-processing. In the zoomed-in view, it can be seen that small areas of noise have been successfully removed and small areas of holes have been reasonably filled.

[0082] Figure 5 It shows the comparison of two different optimization strategies - the mainstream opening and closing operation and the method proposed in the present invention - when processing the high-resolution remote sensing image change detection results. The left side is the result of processing using the mainstream method. Due to its global processing method, it is easy to cause over-processing, resulting in blurred boundaries and the introduction of new artifacts or distortions. The right side is the optimization result obtained by applying the method proposed in the present invention. Through the area-constrained hole filling and edge-constrained reconstruction technology, it not only effectively removes small-area noise, but also reasonably fills small-area holes, while retaining important edge information in the changed area, avoiding edge blur or distortion. By comparison, it can be seen that the high-resolution remote sensing image processing method provided by the present invention is more sophisticated in detail processing than the traditional mainstream method, and can more effectively serve subsequent analysis and application needs.

[0083] In summary, it can be seen that in the embodiment provided by the present invention, the remote sensing image is preprocessed to improve the accuracy and usability of the image. By analyzing the connected components, the hole filling of the area constraint and the reconstruction after denoising, the interference factors in the image are effectively controlled, and the changed area is clearer. At the same time, when expanding, the reconstruction template is used as a constraint to ensure that the expansion reconstruction does not cross the boundary, which can effectively avoid the boundary blur or distortion caused by the expansion phenomenon and retain important edge information.

[0084] The above describes the embodiment of the present invention from the perspective of a method for processing high-resolution remote sensing images. The following describes the embodiment of the present invention from the perspective of a device for processing high-resolution remote sensing images.

[0085] See also Figure 6 , a virtual structural diagram of a high-resolution remote sensing image processing device in an embodiment of the present invention, the high-resolution remote sensing image processing device 600 comprises:

[0086] An acquisition module 601 is used to acquire a front-phase high-resolution remote sensing image and a back-phase high-resolution remote sensing image corresponding to a target image area;

[0087] A preprocessing module 602 is used to preprocess the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0088] A comparison module 603 is used to compare the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0089] An analysis module 604 is used to perform connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, where the region set is a set of independent and mutually unconnected regions in the non-changing area;

[0090] A reconstruction module 605 is used to perform morphological reconstruction on each target region in the target region subset to obtain a hole filling result map, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0091] A denoising module 606 is used to perform a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0092] A construction module 607, configured to construct a reconstruction template covering the target image area based on the initial change detection result map;

[0093] The processing module 608 is used to process the initial change detection result map according to the reconstruction template and the small area noise removal result map to obtain a target change detection result map.

[0094] In a possible design, the analysis module 604 is specifically used for:

[0095] Extracting a non-changing area from the initial change detection result graph to obtain the non-changing area;

[0096] Determining the connectivity number corresponding to the target image area;

[0097] The non-changing region is processed based on the connectivity number to obtain the region set.

[0098] In a possible design, the reconstruction module 605 is specifically used for:

[0099] Taking each target area in the target area subset as a seed point;

[0100] Performing regional expansion and reconstruction according to the seed point and the boundary of each target region, and determining a set of elements that exceed each target region after expansion and reconstruction;

[0101] adding the set of elements to a marked area;

[0102] Repeat the above steps until the elements in the marked area no longer change, and obtain a reconstruction result image;

[0103] The complement of the reconstruction result map is calculated to obtain the hole filling result map.

[0104] In a possible design, the denoising module 606 is specifically used for:

[0105] Construct a target structure element of a preset size;

[0106] Traversing each pixel point in the hole filling result image based on the target structure element to determine the overlap between the image pixels of the hole filling result image and the structure elements of the target structure element;

[0107] The small area noise removal result map is determined according to the overlapping condition.

[0108] In a possible design, the denoising module 606 determines the small area noise removal result image according to the overlapping condition, including:

[0109] Determine, according to the overlapping condition, target image pixels of the structural elements in the hole filling result image that fail to completely cover the target structural element;

[0110] The target image pixels are deleted from the hole filling result map to obtain the small area noise result map.

[0111] In a possible design, the processing module 608 is specifically used for:

[0112] Determine the small area noise removal result image as a seed, and determine the reconstruction template as a constraint condition;

[0113] Dilation is performed based on the seed and the constraint condition to obtain the target change detection result graph.

[0114] In one possible design, the preprocessing includes at least one of radiation correction, geometric correction, and atmospheric correction.

[0115] above Figure 6 The high-resolution remote sensing image processing device in the embodiment of the present invention is described from the perspective of modular functional entities. The high-resolution remote sensing image processing device in the embodiment of the present invention is described in detail from the perspective of hardware processing. Figure 7 , a schematic diagram of an embodiment of a high-resolution remote sensing image processing device 700 in an embodiment of the present invention, the high-resolution remote sensing image processing device 700 comprises:

[0116] An input device 701, an output device 702, a processor 703 and a memory 704 (the number of the processor 703 can be one or more, Figure 7 In some embodiments of the present invention, the input device 701, the output device 702, the processor 703 and the memory 704 may be connected via a communication bus or other means, wherein: Figure 7 The communication bus connection is taken as an example.

[0117] Wherein, by calling the operation instruction stored in the memory 704, the processor 703 is used to perform the following steps:

[0118] Obtaining the previous phase high-resolution remote sensing image and the next phase high-resolution remote sensing image corresponding to the target image area;

[0119] Preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0120] Comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0121] Performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing area;

[0122] Performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result image, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0123] Performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0124] Constructing a reconstruction template covering the target image area based on the initial change detection result map;

[0125] The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

[0126] By calling the operation instructions stored in the memory 704, the processor 703 is also used to execute Figure 1 Any method in the corresponding embodiment.

[0127] See also Figure 8 , Figure 8 A schematic diagram of an electronic device according to an embodiment of the present invention.

[0128] like Figure 8 As shown, an embodiment of the present invention provides an electronic device, including a memory 810, a processor 820, and a computer program 811 stored in the memory 810 and executable on the processor 820. When the processor 820 executes the computer program 811, the following steps are implemented:

[0129] Obtaining the previous phase high-resolution remote sensing image and the next phase high-resolution remote sensing image corresponding to the target image area;

[0130] Preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0131] Comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0132] Performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing area;

[0133] Performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result image, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0134] Performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0135] Constructing a reconstruction template covering the target image area based on the initial change detection result map;

[0136] The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

[0137] In the specific implementation process, when the processor 820 executes the computer program 811, it can achieve Figure 1 Any implementation manner in the corresponding embodiments.

[0138] Since the electronic device introduced in this embodiment is a device used to implement a high-resolution remote sensing image processing device in the embodiment of the present invention, based on the method introduced in the embodiment of the present invention, technical personnel in this field can understand the specific implementation method of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present invention will not be described in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present invention falls within the scope of protection of the present invention.

[0139] See also Fig. 9 , Fig. 9 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention.

[0140] like Fig. 9 As shown, the embodiment of the present invention further provides a computer-readable storage medium 900, on which a computer program 911 is stored. When the computer program 911 is executed by a processor, the following steps are implemented:

[0141] Obtaining the previous phase high-resolution remote sensing image and the next phase high-resolution remote sensing image corresponding to the target image area;

[0142] Preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image;

[0143] Comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map;

[0144] Performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing area;

[0145] Performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result image, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold;

[0146] Performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image;

[0147] Constructing a reconstruction template covering the target image area based on the initial change detection result map;

[0148] The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

[0149] In the specific implementation process, the computer program 911 is executed by the processor to achieve Figure 1 Any implementation manner in the corresponding embodiments.

[0150] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0151] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0155] The embodiment of the present invention also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the following Figure 1 The process in the corresponding embodiment.

[0156] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0157] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0158] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There are other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0159] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0161] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0162] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing high-resolution remote sensing images, characterized in that: include: Obtaining the previous phase high-resolution remote sensing image and the next phase high-resolution remote sensing image corresponding to the target image area; Preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image; Comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map; Performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing area; Performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result image, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold; Performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image; Constructing a reconstruction template covering the target image area based on the initial change detection result map; The initial change detection result map is processed according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map.

2. The method according to claim 1, characterized in that The performing connected component analysis on the non-changing area corresponding to the initial change detection result graph to obtain the area set includes: Extracting a non-changing area from the initial change detection result graph to obtain the non-changing area; Determining the connectivity number corresponding to the target image area; The non-changing region is processed based on the connectivity number to obtain the region set.

3. The method according to claim 1, characterized in that The performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result map comprises: Taking each target area in the target area subset as a seed point; Performing regional expansion and reconstruction according to the seed point and the boundary of each target region, and determining a set of elements that exceed each target region after expansion and reconstruction; adding the set of elements to a marked area; Repeat the above steps until the elements in the marked area no longer change, and obtain a reconstruction result image; The complement of the reconstruction result map is calculated to obtain the hole filling result map.

4. The method according to claim 1, characterized in that: The performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image includes: Construct a target structure element of a preset size; Traversing each pixel point in the hole filling result image based on the target structure element to determine the overlap between the image pixels of the hole filling result image and the structure elements of the target structure element; The small area noise removal result map is determined according to the overlapping condition.

5. The method according to claim 4, characterized in that Determining the small area noise removal result map according to the overlapping situation includes: Determine, according to the overlapping condition, target image pixels of the structural elements in the hole filling result image that fail to completely cover the target structural element; The target image pixels are deleted from the hole filling result map to obtain the small area noise result map.

6. The method according to any one of claims 1 to 5, characterized in that The processing of the initial change detection result map according to the reconstructed template and the small area noise removal result map to obtain a target change detection result map includes: Determine the small area noise removal result image as a seed, and determine the reconstruction template as a constraint condition; Dilation is performed based on the seed and the constraint condition to obtain the target change detection result graph.

7. The method according to any one of claims 1 to 5, characterized in that The preprocessing includes at least one of radiation correction, geometric correction and atmospheric correction.

8. A high-resolution remote sensing image processing device, characterized in that: include: An acquisition module is used to acquire a front-phase high-resolution remote sensing image and a back-phase high-resolution remote sensing image corresponding to a target image area; A preprocessing module, used for preprocessing the front-phase high-resolution remote sensing image and the back-phase high-resolution remote sensing image to obtain a first remote sensing image and a second remote sensing image; A comparison module, used for comparing the first remote sensing image with the second remote sensing image to obtain an initial change detection result map; An analysis module, configured to perform connected component analysis on the non-changing region corresponding to the initial change detection result graph to obtain a region set, wherein the region set is a set of independent and mutually unconnected regions in the non-changing region; A reconstruction module, used for performing morphological reconstruction on each target region in the target region subset to obtain a hole filling result map, wherein the target region subset is a set of regions in the region set whose area is smaller than a set threshold; A denoising module, used for performing a small area noise removal operation on the hole filling result image to obtain a small area noise removal result image; A construction module, used for constructing a reconstruction template covering the target image area based on the initial change detection result map; A processing module is used to process the initial change detection result map according to the reconstruction template and the small area noise removal result map to obtain a target change detection result map.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to implement the high-resolution remote sensing image processing method as described in any one of claims 1 to 7 when executing a computer management program stored in the memory.

10. A computer-readable storage medium having a computer management program stored thereon, characterized in that: When the computer management program is executed by the processor, the high-resolution remote sensing image processing method as described in any one of claims 1 to 7 is implemented.

Images