Image data processing method for surveying and mapping

Abstract

The invention discloses an image data processing method for surveying and mapping, and particularly relates to the field of image data processing, which comprises the following steps: acquiring optical image data and near-infrared data of a surveying and mapping building area, extracting core features for accurate mapping, and establishing background element spectrum contrast features; generating a background mask based on a comparison result of the spectral features and background element spectral contrast features, and segmenting a background region in combination with a region growing dynamic segmentation model; marking the grids which are not completely marked with the background masks in the orthographic image as a building background mixed area, carrying out background mask secondary marking on the building background mixed area, and carrying out background feature suppression; extracting a preliminary contour of the building by adopting an edge detection algorithm, and performing curve multi-level progressive fitting on the extracted contour edge of the building; and performing element calibration and data layer superposition on the building contour based on curve fitting to form comprehensive expression of the building and geographic feature elements.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and more specifically, to an image data processing method for surveying and mapping. Background Art

[0002] In the field of architectural surveying and object segmentation, the accurate distinction between buildings and complex backgrounds (such as vegetation, water bodies, shadows, etc.) is a long-standing technical difficulty. Complex background areas are often easily confused with buildings due to the high similarity of texture, shape and spectral characteristics, resulting in limitations in the accurate extraction of building areas by traditional segmentation methods. Especially in densely populated urban areas or natural environments, buildings often overlap with background elements, such as vegetation covering part of the building or water reflection interfering with the edge features of the building, making the distinction even more difficult.

[0003] Near-infrared (NIR) data has a strong ability to distinguish background elements such as vegetation and water bodies in terms of reflectivity. For example, vegetation has high reflectivity in the NIR band, while buildings have low reflectivity. However, segmentation methods that rely solely on near-infrared data often show insufficient accuracy when dealing with scenes with similar spectral characteristics or complex textures. In addition, the transition areas between building boundaries and backgrounds in complex backgrounds (such as parts of buildings obscured by vegetation) are usually inaccurately or mislabeled, and existing methods lack the ability to process these areas in a refined manner.

[0004] How to propose an image data processing method that suppresses complex background features and combines edge detection and curve fitting technology to achieve accurate mapping is an urgent problem that needs to be solved.

[0005] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an image data processing method for surveying and mapping to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] S1: Collect and extract the core features of the optical image data and near-infrared data of the surveyed building area, accurately map the core features in two dimensions to form a plane grid structure, measure the near-infrared reflectance spectrum characteristic data of non-building elements, and establish the spectral comparison features of background elements;

[0009] S2: Extract the spectral features in the grid, generate a background mask based on the comparison results of the spectral features and the spectral comparison features of the background elements, and segment the background area in combination with the dynamic segmentation model of region growing;

[0010] S3: Mark the grids in the orthophoto image that are not fully annotated with the background mask as the building background mixed area, perform secondary annotation of the building background mixed area with the background mask, and suppress the background features of the secondary annotated background mask area;

[0011] S4: Using the Canny edge detection algorithm to extract the preliminary outline of the building, and performing multi-level progressive fitting of the B-Spline curve on the extracted building outline edge;

[0012] S5: The building outline after B-Spline curve fitting is calibrated and superimposed on the data layer to form a comprehensive expression of the building and geographical feature elements.

[0013] In a preferred embodiment, in S1, collecting and extracting the core features of the optical image data and near-infrared data of the surveyed building area, accurately mapping the core features in two dimensions to form a plane grid structure, measuring the near-infrared reflectance spectrum characteristic data of non-building elements, and establishing the background element spectrum comparison features specifically include:

[0014] Collect multimodal data of surveyed and mapped building areas through high-precision remote sensing platforms, including optical image data and near-infrared data;

[0015] Through multi-view image matching and geographic calibration, all modal data are unified to the same coordinate system and projection standard to form an orthophoto, and the core features of the optical image data and near-infrared data in the modal data are extracted and aligned;

[0016] The core features of the aligned different modalities are accurately mapped in two dimensions in the orthoimage to form a plane grid structure, in which each grid contains at least one pixel point;

[0017] Analyze the grids after modal feature mapping one by one and mark the core features of the grids. The core features include the RGB and texture features of the building extracted by the optical image channel and the reflectance spectrum features of the near-infrared channel.

[0018] Based on the multi-layer feature superposition strategy, RGB, texture features and spectral features are integrated into multi-modal combined features, where each mode represents a corresponding physical feature attribute expression;

[0019] A grid or grid group including all non-building object elements is selected, the near-infrared reflectance spectral characteristic data of the non-building elements is measured, and the spectral comparison characteristics of the background elements are established.

[0020] In a preferred embodiment, in S2, extracting spectral features in the grid, generating a background mask based on the comparison result of the spectral features with the spectral comparison features of the background elements, and segmenting the background region in combination with the dynamic segmentation model of region growing specifically includes:

[0021] Select at least one spectral feature comparison point in each grid, and identify and generate a preliminary background mask of all non-building areas in the image based on the hit of the spectral feature comparison point in the spectral comparison feature of the background element;

[0022] A grid with dense preliminary background mask annotations is selected as the initial point. The adjacent grids are recursively compared using the multimodal combined features of optical image and near-infrared data. Background mask diffusion annotation is performed based on feature similarity, where feature similarity is the feature cosine similarity value that integrates texture, brightness and spectral reflectance.

[0023] In a preferred embodiment, in S3, marking the grids in the orthophoto image that are not fully annotated with the background mask as the building background mixed area, performing secondary annotation of the building background mixed area with the background mask, and performing background feature suppression on the secondary annotated background mask area specifically includes:

[0024] Obtain the grid combination with the unlabeled background mask area in the orthophoto image, mark it as the building background mixed area, and extract the multimodal combination features of the optical image and near-infrared data of all the grids in the building background mixed area;

[0025] Based on the hit results of the spectral features of all pixels in the grid in the spectral comparison features of the background elements, the background mask is secondary marked on the mixed area of ​​the building background;

[0026] Based on the feature similarity when the spectral feature hits the spectral comparison feature of the background element, the multi-modal combination feature of the background mask secondary annotation pixel position is proportionally attenuated to weaken the feature expression of the background in the mixed area of ​​the building background;

[0027] The orthophoto image after background feature suppression is used as the initial analysis image of the buildings in the surveying area for contour fitting.

[0028] In a preferred embodiment, in S4, the preliminary outline of the building is extracted using the Canny edge detection algorithm, and the B-Spline curve multi-level progressive fitting is performed on the extracted building outline edge, specifically including:

[0029] The Canny edge detection algorithm is used to extract the edges of the initial analysis images of the buildings in the survey area, construct structured scene contour information, and introduce multi-scale convolution feature extraction to identify the local details and global morphology of the buildings;

[0030] According to the length and complexity of the original outline of the building, the original outline is simplified to obtain key nodes, and the key nodes of the parts with higher geometric complexity are extracted and marked as sparse control point sets, and these sparse control point sets are used to construct a preliminary B-Spline curve;

[0031] The constructed preliminary B-Spline curve is segmented, and the error between the local curvature distribution of each curve segment and the adjacent key nodes in the curve segment interval is calculated. Based on the error calculation results, the threshold comparison method is used to determine whether the detail fitting degree of the curve in this area meets the standard;

[0032] In the curve segments where the detail fitting degree does not meet the standard, the number of control points is increased through control point interpolation, and multi-level cyclic fitting of the B-Spline curve is performed. The fitting of each layer adds an equal number of control points on the basis of the previous layer, and at the same time improves the order of the B-Spline curve. The cycle is terminated until the detail fitting degree of all segments of the B-Spline curve meets the standard, and the fitted building contour edge is obtained.

[0033] In a preferred embodiment, in S5, the building outline after B-Spline curve fitting is subjected to element calibration and data layer superposition to form a comprehensive expression of the building and geographical feature elements, which specifically includes:

[0034] Convert the coordinate data corresponding to the building outline into the building's element information to establish a geographic data layer;

[0035] All fitted building outline boundaries are transformed into a unified coordinate projection and aligned with the spatial position. At the same time, the geographic data layer is superimposed to add attribute information to the building objects to form a comprehensive expression of buildings and geographic feature elements.

[0036] The technical effects and advantages of the image data processing method for surveying and mapping of the present invention are as follows:

[0037] Through the feature extraction and fusion analysis of optical images and near-infrared data, the texture and shape advantages of optical images and the spectral difference characteristics of near-infrared data are effectively utilized to achieve the preliminary separation of buildings and backgrounds. The background mask generated based on spectral feature comparison combined with the regional growth dynamic segmentation model can accurately identify and suppress the interference of complex background areas (such as vegetation and water bodies), especially in the transition area of ​​the building boundary. The mixed feature expression of the complex background and building intersection area is further weakened through secondary annotation and background feature suppression. Finally, through Canny edge detection and B-Spline curve fitting technology, the building outline boundary is gradually optimized at multiple levels, making the shape expression of the building smoother and more realistic, providing high-quality outline data support for subsequent surveying and mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The present invention is a schematic diagram of an image data processing method for surveying and mapping. DETAILED DESCRIPTION

[0039] 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Example 1

[0041] Figure 1 The present invention provides an image data processing method for surveying and mapping, which comprises the following steps:

[0042] S1: Collect and extract the core features of the optical image data and near-infrared data of the surveyed building area, accurately map the core features in two dimensions to form a plane grid structure, measure the near-infrared reflectance spectrum characteristic data of non-building elements, and establish the spectral comparison features of background elements;

[0043] S2: Extract the spectral features in the grid, generate a background mask based on the comparison results of the spectral features and the spectral comparison features of the background elements, and segment the background area in combination with the dynamic segmentation model of region growing;

[0044] S3: Mark the grids in the orthophoto image that are not fully annotated with the background mask as the building background mixed area, perform secondary annotation of the building background mixed area with the background mask, and suppress the background features of the secondary annotated background mask area;

[0045] S4: Using the Canny edge detection algorithm to extract the preliminary outline of the building, and performing multi-level progressive fitting of the B-Spline curve on the extracted building outline edge;

[0046] S5: The building outline after B-Spline curve fitting is calibrated and superimposed on the data layer to form a comprehensive expression of the building and geographical feature elements.

[0047] In S1, the core features of the optical image data and near-infrared data of the surveyed building area are collected and extracted, the core features are accurately mapped in two dimensions to form a plane grid structure, the near-infrared reflectance spectral characteristic data of non-building elements are measured, and the spectral comparison features of background elements are established.

[0048] In the task of building area mapping, high-precision remote sensing platforms (such as drones or satellites) are used to collect optical image data and near-infrared (NIR) data. Optical image data is used to obtain the texture, shape and color characteristics of buildings. The images collected from multiple perspectives are aligned through multi-perspective image matching algorithms (such as SIFT key point matching), and the multi-modal data are unified to the same geographic coordinate system and projection standard using geographic calibration technology to generate high-precision orthophotos, ensuring the spatial consistency and registration accuracy of multi-modal data.

[0049] Core feature extraction is performed on the optical image and near-infrared data in the orthophoto. RGB features (such as color histogram) and texture features (such as gray-level co-occurrence matrix, LBP features) are extracted from the optical image; reflectance spectrum characteristics are extracted from the near-infrared data. In view of the differences in resolution and feature distribution of different modal data, the low-resolution near-infrared data is super-reconstructed using a resolution matching algorithm, and pixel-level feature alignment technology is used to ensure that the optical and near-infrared modal features accurately correspond in the same spatial grid.

[0050] The core features of the aligned optical image and near-infrared data are accurately mapped in two dimensions in the orthoimage, and a unified plane grid structure is generated through a pixel-level multimodal feature overlay strategy. Each grid contains the fusion features of the optical image and near-infrared data, and contains at least one pixel. The fusion features include RGB color, texture features, and spectral features. The features of each modality are represented by multiple channels to support subsequent analysis.

[0051] The grids after modal feature mapping are analyzed grid by grid, and the core features of the grid are annotated. The core features include the RGB and texture features of the building extracted by the optical image channel, and the reflectance spectrum features of the near-infrared channel. The RGB features extract the color features of the optical image in the grid, which are used to represent the visual information of the building or background area. The texture features extract the edge and structure information of the optical image to enhance the description of the shape of the building. The spectral features annotate the reflectance spectrum characteristics of the near-infrared data to distinguish between the background (such as vegetation and water bodies) and the building.

[0052] Select samples in the grid or grid group that completely contain non-building object elements (such as vegetation, water bodies, and bare soil), and establish the spectral comparison characteristics of the background elements by analyzing their near-infrared spectral reflectance characteristics. The specific steps are as follows:

[0053] Background sample screening: based on the color characteristics of optical images and the spectral distribution of near-infrared data, highly representative background grids (such as vegetation areas with high NIR reflectivity and water areas with low reflectivity) are screened out;

[0054] Spectral data measurement: calculate the reflection spectrum curve of the screened background samples and record the reflection characteristics of the background elements in the near-infrared band;

[0055] The spectral comparison feature is established, and the spectral characteristics of the background samples are clustered (such as K-Means clustering) to generate a spectral comparison database of background elements for subsequent background distinction and mask generation.

[0056] In S2, the spectral features in the grid are extracted, and a background mask is generated based on the comparison results of the spectral features with the spectral comparison features of the background elements. The background area is segmented by combining the dynamic segmentation model of region growing.

[0057] In each grid, at least one spectral feature point is extracted from the near-infrared band reflectance characteristics, and the pixel points with the most representative reflectance characteristics are selected first to ensure that the spectral characteristics of the grid can be accurately expressed. For areas with complex spectral characteristics, multiple pixels can be selected to improve the expression accuracy.

[0058] The spectral features of each comparison point are matched with the pre-established background element spectral reference feature library. The matching degree between the comparison point and the background spectral features is calculated using Euclid, and the features whose matching degree exceeds the set standard are marked as feature hits.

[0059] From the preliminary background mask, grid groups with high background annotation density are selected as initial points. These grid groups have high background confidence and can be used as the starting area for diffuse annotation. The selection criteria include that the background mask annotation ratio exceeds the set threshold (80%) and the cosine similarity of the spectral feature hitting the background control feature is higher than the set threshold (0.85). The multimodal combination features of optical image and near-infrared data are used to recursively compare adjacent grids, and background mask diffuse annotation is performed based on feature similarity, where feature similarity is the feature cosine similarity value of the fused texture, brightness and spectral reflectance. The similarity value calculation formula is:

[0060]

[0061] Where, X i is the eigenvector of the i-th initial point, Y i is the feature vector of the ith adjacent grid, n is the number of initial points set, when a set of X i With Y i After the comparison, Y i Automatically converted to X i Perform the next adjacent grid comparison, and for the newly marked background grid, repeat the multimodal feature extraction and adjacent grid comparison process to achieve the gradual diffusion annotation of the mask.

[0062] In S3, the grids in the orthophoto image that are not fully annotated with the background mask are marked as building background mixed areas, the building background mixed areas are secondary annotated with the background mask, and the background features are suppressed in the secondary annotated background mask areas.

[0063] Identify areas with unlabeled background masks from orthophotos. These areas are usually located in the transition area between buildings and backgrounds, such as the edge of buildings partially blocked by vegetation. By analyzing the multimodal mapping images of optical images and near-infrared data, identify these unlabeled areas (areas that failed to diffuse during recursive diffusion annotation) as building-background mixed areas. These areas not only have building characteristics, but are also affected by the surrounding complex background (such as vegetation, shadows, water bodies, etc.), and usually contain spectral and texture characteristics intertwined with buildings and backgrounds. Extract multimodal combined features of optical images and near-infrared data from building-background mixed areas.

[0064] The spectral features of each grid pixel in the mixed area of ​​the building background are compared, and the spectral feature hit detection is performed on the pixels in each grid using the established background element spectral comparison feature library. The cosine similarity of the features is calculated (also used as the feature attenuation ratio), and the degree of match between the spectral features of each pixel and the background element features is determined. Features whose matching degree exceeds the set standard are marked as feature hits. According to the spectral feature hit results, the pixels in the mixed area of ​​the building background are annotated with a secondary background mask. For pixels with higher spectral similarity, they are marked as background areas. For pixels with lower similarity, they are retained as building areas. Through this step, the transition area between the building and the background is accurately separated to ensure the accuracy of the mask.

[0065] Based on cosine similarity, the multimodal combined features of the background mask pixels after secondary annotation are weighted attenuated. Since the mixed area of ​​the building background contains the spectral features of complex background elements, it is necessary to weaken the feature expression of the background area. The feature attenuation strategy is adopted to reduce the weight of the background spectral features in the combined features, thereby weakening the influence of the background elements.

[0066] For the texture and light and shadow features of the background area, low-pass filtering or other attenuation methods are used to reduce the interference of the background on the building. The orthophoto image after background feature suppression is used as the initial analysis image of the building in the surveying area for contour fitting.

[0067] In S4, the Canny edge detection algorithm is used to extract the preliminary outline of the building, and the extracted building outline edge is subjected to multi-level progressive fitting of the B-Spline curve.

[0068] The Canny edge detection algorithm is used to extract the edges of the building images in the survey area. The Canny algorithm uses a multi-stage process, first performing Gaussian filtering to remove noise, then calculating the gradient strength and direction, using non-maximum suppression to refine the edges, and finally performing double threshold detection to extract all potential edges in the image, especially the outer edges of the buildings. This step can effectively highlight the outline of the building and ignore background noise.

[0069] In order to further improve the recognition ability of building details, multi-scale convolution feature extraction is introduced. The building image is convolved by convolution kernels of different scales to extract the local details and global morphological features of the image. This method can obtain the global outline and local complex shapes of the building at different scales, and enhance the recognition ability of complex structures of buildings, especially in edge details and complex roofs.

[0070] Based on the results of Canny edge detection and multi-scale feature extraction, the original outline of the building is simplified. By fitting polygons or smoothing curves, redundant edge points and details are removed to obtain a relatively concise preliminary outline. This simplification process focuses on the structural characteristics of the outline while retaining the basic characteristics of the building's morphology.

[0071] Based on the length and complexity of the original building outline, representative key nodes are extracted by analyzing the geometric form of the outline. Generally, the location and distribution of nodes reflect the complexity and geometry of the building outline. According to the geometric complexity of the outline, the parts with large local geometric changes are screened out and marked as "sparse control points". The specific labeling strategy of the control points is to calculate the contour curvature (second derivative) and find the parts with large curvature changes as control points. The expression is:

[0072]

[0073] Where κ is the curvature (used to represent set complexity), x(t) and y(t) are the coordinates of the contour points, t is the actual parameter of the contour coordinates, and x ′ (t), x″(t), and y ′ (t), y″(t) are the first and second order derivatives of x(t) and y(t), respectively.

[0074] These sparse control point sets will serve as the basis for the subsequent B-Spline curve construction. Use these sparse control point sets to construct a preliminary B-Spline curve, which is defined as:

[0075]

[0076] Where C(μ) is the B-Spline curve, N j,k(μ) is the B-Spline basis function, P j is the control point, k is the order of B-Spline, μ is a real number in the parameter domain, which determines the shape and position of the curve. The specific interval depends on the settings of the control point and B-Spline basis function.

[0077] The preliminary B-Spline curve is processed in segments, and the local curvature distribution of each segment is calculated. The curvature reflects the degree of bending of the curve, and the height of the local curvature can reflect the complexity of the building's form. Further, the error of each curve segment is calculated, that is, the distance error between each control point and its adjacent key nodes in the local area. This error value will be used as a standard for evaluating the accuracy of curve fitting. Based on the error calculation results, the threshold comparison method is used to judge the degree of detail fitting of each curve segment. If the error exceeds the preset tolerance threshold (0.5 pixels), it indicates that the degree of detail fitting in this area does not meet the standard and needs further optimization. For curve segments that meet the fitting accuracy requirements, the existing fitting structure is maintained.

[0078] For curve segments with unsatisfactory fitting accuracy, the number of control points is increased through control point interpolation to improve the fitting accuracy of the curve. Each time interpolation is performed, new control points are added equidistantly on the basis of the current curve segment, and the density of control points is gradually increased to refine the shape of the curve. On this basis, the order of the B-Spline curve can also be increased (from k to k+1) to make the curve fit the details of the building more accurately.

[0079] The process of adding control points and increasing the order is repeated in multiple levels until the detail fitting degree of all curve segments reaches the set standard. Each layer of fitting is performed on the basis of the previous layer, gradually optimizing the details of the contour to ensure the accuracy and smoothness of the building contour. When the detail fitting of all B-Spline curve segments reaches the predetermined accuracy requirement, the cycle is terminated. At this point, the obtained B-Spline curve accurately represents the shape of the building contour.

[0080] In S5, the building outline after B-Spline curve fitting is calibrated and the data layer is superimposed to form a comprehensive expression of buildings and geographical features.

[0081] The spatial coordinates of the building's outline in the two-dimensional coordinate system are converted into coordinate data in the geographic coordinate system. In order to make these coordinates compatible with the existing geographic data layer, a coordinate projection transformation is used, such as converting from the geographic coordinate system (WGS84) to the local plane coordinate system. The building's outline coordinate data is combined with related feature information to construct the attribute information of the building element. For example, the building's height, purpose, structure type, etc. can be part of the feature information. This information is associated with the outline coordinate data of each building to form a feature set containing spatial and attribute data.

[0082] All fitted building outline coordinates are uniformly transformed into coordinate projections, so that the coordinates of different data sources are aligned in the same coordinate system. The coordinate values ​​of each building are converted from the original space to the target space through the projection transformation algorithm. By comparing the geographic coordinate information of different data sources, it is ensured that the building outlines are correctly matched in space to avoid geometric inconsistencies caused by errors or offsets.

[0083] Finally, the building outline data will be superimposed on the existing geographic data layer. The geographic data layer includes geographic feature data such as terrain, roads, vegetation, and water bodies, which can be superimposed through GIS (Geographic Information System). The spatial location and outline boundaries of the building will be aligned with the geographic data layer to form a unified comprehensive data layer that supports spatial analysis and image visualization.

[0084] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0085] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can 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. The semiconductor medium can be a solid-state hard disk.

[0086] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0087] 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 modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0088] In the several embodiments provided in the present application, 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 modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules 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 through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0089] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0091] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several 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 methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program codes.

[0092] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0093] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for processing image data for surveying and mapping, characterized in that: The steps include: S1: Collect and extract the core features of the optical image data and near-infrared data of the surveyed building area, accurately map the core features in two dimensions to form a plane grid structure, measure the near-infrared reflectance spectrum characteristic data of non-building elements, and establish the spectral comparison features of background elements; S2: Extract the spectral features in the grid, generate a background mask based on the comparison results of the spectral features and the spectral comparison features of the background elements, and segment the background area in combination with the dynamic segmentation model of region growing; S3: Mark the grids in the orthophoto image that are not fully annotated with the background mask as the building background mixed area, perform secondary annotation of the building background mixed area with the background mask, and suppress the background features of the secondary annotated background mask area; S4: Using the Canny edge detection algorithm to extract the preliminary outline of the building, and performing multi-level progressive fitting of the B-Spline curve on the extracted building outline edge; S5: The building outline after B-Spline curve fitting is calibrated and superimposed on the data layer to form a comprehensive expression of the building and geographical feature elements.

2. The image data processing method for surveying and mapping according to claim 1, characterized in that: In S1, the core features of the optical image data and near-infrared data of the surveyed building area are collected and extracted, the core features are accurately mapped in two dimensions to form a plane grid structure, the near-infrared reflectance spectrum characteristic data of non-building elements are measured, and the spectral comparison features of the background elements are established. Specifically, the following are included: Collect multimodal data of surveyed and mapped building areas through high-precision remote sensing platforms, including optical image data and near-infrared data; Through multi-view image matching and geographic calibration, all modal data are unified to the same coordinate system and projection standard to form an orthophoto, and the core features of the optical image data and near-infrared data in the modal data are extracted and aligned; The core features of the aligned different modalities are accurately mapped in two dimensions in the orthoimage to form a plane grid structure, in which each grid contains at least one pixel point; Analyze the grids after modal feature mapping one by one and mark the core features of the grids. The core features include the RGB and texture features of the building extracted by the optical image channel and the reflectance spectrum features of the near-infrared channel. Based on the multi-layer feature superposition strategy, RGB, texture features and spectral features are integrated into multi-modal combined features, where each mode represents a corresponding physical feature attribute expression; A grid or grid group including all non-building object elements is selected, the near-infrared reflectance spectral characteristic data of the non-building elements is measured, and the spectral comparison characteristics of the background elements are established.

3. The image data processing method for surveying and mapping according to claim 2, characterized in that: In S2, the spectral features in the grid are extracted, and a background mask is generated based on the comparison results of the spectral features and the spectral comparison features of the background elements. The background area is segmented by combining the dynamic segmentation model of region growing, which specifically includes: Select at least one spectral feature comparison point in each grid, and identify and generate a preliminary background mask of all non-building areas in the image based on the hit of the spectral feature comparison point in the spectral comparison feature of the background element; A grid with dense preliminary background mask annotations is selected as the initial point. The adjacent grids are recursively compared using the multimodal combined features of optical image and near-infrared data. Background mask diffusion annotation is performed based on feature similarity, where feature similarity is the feature cosine similarity value that integrates texture, brightness and spectral reflectance.

4. The image data processing method for surveying and mapping according to claim 3, characterized in that: In S3, the grids in the orthophoto image that are not fully annotated with the background mask are marked as the building background mixed area, the building background mixed area is secondary annotated with the background mask, and the background feature suppression is performed on the secondary annotated background mask area. Specifically, the following steps are performed: Obtain the grid combination with the unlabeled background mask area in the orthophoto image, mark it as the building background mixed area, and extract the multimodal combination features of the optical image and near-infrared data of all the grids in the building background mixed area; Based on the hit results of the spectral features of all pixels in the grid in the spectral comparison features of the background elements, the background mask is secondary marked on the mixed area of ​​the building background; Based on the feature similarity when the spectral feature hits the spectral comparison feature of the background element, the multi-modal combination feature of the background mask secondary annotation pixel position is proportionally attenuated to weaken the feature expression of the background in the mixed area of ​​the building background; The orthophoto image after background feature suppression is used as the initial analysis image of the buildings in the surveying area for contour fitting.

5. The image data processing method for surveying and mapping according to claim 4, characterized in that: In S4, the Canny edge detection algorithm is used to extract the preliminary outline of the building, and the B-Spline curve multi-level progressive fitting is performed on the extracted building outline edge, specifically including: The Canny edge detection algorithm is used to extract the edges of the initial analysis images of the buildings in the survey area, construct structured scene contour information, and introduce multi-scale convolution feature extraction to identify the local details and global morphology of the buildings; According to the length and complexity of the original outline of the building, the original outline is simplified to obtain key nodes, and the key nodes of the parts with higher geometric complexity are extracted and marked as sparse control point sets, and these sparse control point sets are used to construct a preliminary B-Spline curve; The constructed preliminary B-Spline curve is segmented, and the error between the local curvature distribution of each curve segment and the adjacent key nodes in the curve segment interval is calculated. Based on the error calculation results, the threshold comparison method is used to determine whether the detail fitting degree of the curve in this area meets the standard; In the curve segments where the detail fitting degree does not meet the standard, the number of control points is increased through control point interpolation, and multi-level cyclic fitting of the B-Spline curve is performed. The fitting of each layer adds an equal number of control points on the basis of the previous layer, and at the same time improves the order of the B-Spline curve. The cycle is terminated until the detail fitting degree of all segments of the B-Spline curve meets the standard, and the fitted building contour edge is obtained.

6. The image data processing method for surveying and mapping according to claim 3, characterized in that: In S5, the building outline after B-Spline curve fitting is calibrated and the data layer is superimposed to form a comprehensive expression of buildings and geographical features, including: Convert the coordinate data corresponding to the building outline into the building's element information to establish a geographic data layer; All fitted building outline boundaries are transformed into a unified coordinate projection and aligned with the spatial position. At the same time, the geographic data layer is superimposed to add attribute information to the building objects to form a comprehensive expression of buildings and geographic feature elements.

Images