Historical topographic map geographic element identification method based on image features

Abstract

The invention provides a historical topographic map geographic element identification method based on image features, and belongs to the technical field of image identification, and the method comprises the steps: carrying out the preprocessing of a historical topographic map, and carrying out the conversion of a color model; performing color clustering analysis on the preprocessed map image based on color features to obtain images of ground features of different colors after clustering segmentation; the topographic map image is converted into a grey-scale map, and a gray-scale co-occurrence matrix is adopted to extract texture features of the image; the texture features are visualized; and the geographic elements on the historical topographic map are identified in combination with the color features and the texture features. According to the method, the geographic element recognition effect can be visually displayed, the recognition efficiency and accuracy are improved, and actual technical support can be provided for historical geographic research.

Description

[0001] A method for identifying geographical elements in historical topographic maps based on image features Technical Field

[0003] The present invention relates to the technical field of image recognition, and in particular to a method for identifying geographical elements of historical topographic maps based on image features. Background Art

[0005] As an important document recording geographical changes and cultural heritage, historical topographic maps contain rich historical information and are of great significance for studying land use changes, the impact of human activities on the environment, and the long-term interaction between humans and the natural environment. These maps not only provide valuable data resources for multiple disciplines such as geography, archaeology, and history, but also provide empirical basis and reference basis for related research fields. With the continuous development of science and technology, the extraction and identification methods of geographical elements in historical topographic maps are also being continuously optimized and innovated.

[0006] At present, the extraction and recognition methods of geographic elements mainly rely on advanced technologies such as geographic information systems, computer vision and pattern recognition. These technologies have effectively solved the problems of noise, color confusion and poor image quality in images to a certain extent, and improved the accuracy of geographic element recognition. However, although the existing technologies have made significant progress in some aspects, they still face problems such as huge computational complexity, lengthy processing time and insufficient classification accuracy when processing large-format color historical topographic maps. These problems directly limit the efficiency and accuracy of geographic element recognition in historical topographic maps. In addition, the extraction of line elements faces many severe challenges. Specifically, there is a significant interlacing and overlapping phenomenon between line elements and background plain dye elements, and the two often have high similarity in color and energy information, which makes it particularly complicated to accurately extract line elements. In this case, the application effect of existing technologies when processing line elements with complex backgrounds often fails to meet expectations. This limitation emphasizes the urgent need for targeted improvements to existing methods to develop more efficient and accurate recognition algorithms, thereby improving the overall performance and reliability of line element extraction.

[0007] Based on the above problems, a new method for identifying geographical elements in historical topographic maps is urgently needed to improve the efficiency and accuracy of identifying geographical elements in historical maps. Summary of the invention

[0009] The purpose of the present invention is to propose a method for identifying geographical elements in historical topographic maps based on image features, which identifies geographical elements in topographic maps according to their color features and texture features, and provides a reference for exploring the changes of geographical elements over time.

[0010] The technical solution of the present invention is as follows:

[0011] Historical topographic map preprocessing;

[0012] Perform color clustering analysis on the preprocessed map image based on color features to obtain images of objects of different colors after cluster segmentation;

[0013] The gray-level co-occurrence matrix is ​​used to extract the texture features of the image;

[0014] Visualize texture features;

[0015] Combine color features and texture features to identify geographical features on historical maps.

[0016] According to a preferred embodiment, the preprocessing of the historical map includes converting the historical topographic map from RGB color space to HSV color space to enhance the accuracy of color clustering and reshaping the three-dimensional HSV image array into a two-dimensional array for clustering.

[0017] According to a preferred embodiment, the map image is subjected to color clustering, comprising:

[0018] Construct pixel coordinates based on the hue, saturation, and brightness values ​​of the pixel;

[0019] Determine the number of cluster centers based on the image color category;

[0020] Randomly select the initial cluster center in the pixel coordinates;

[0021] Calculate the distance d between each pixel and the cluster center using the following formula:

[0022] d=

[0023] in , They are the coordinates of the cluster center and the current calculated pixel point respectively;

[0024] Assign pixels to the nearest cluster centers to form clusters with the same number of cluster centers;

[0025] The arithmetic mean of the hue, saturation, and brightness of all pixels in each cluster is used as the new cluster center, and the position of the cluster center is updated until the change in the cluster center is less than 0.0001.

[0026] According to a preferred embodiment, the segmented and clustered images of objects of different colors include:

[0027] Assign a label to each cluster to distinguish different clusters;

[0028] Iterate over each cluster and create a segmented color image for each cluster.

[0029] According to a preferred embodiment, it is characterized in that the extraction of texture features of an image using a gray level co-occurrence matrix comprises:

[0030] Convert the color historical topographic map into grayscale image to eliminate the redundant color information in the original image;

[0031] Compute the gray-level co-occurrence matrix to obtain statistical information about pixel pairs in the image;

[0032] The texture feature values ​​are extracted from the gray-level co-occurrence matrix, where contrast represents the clarity of the image texture; correlation represents the directionality of the image texture; energy represents the roughness of the image texture; entropy represents the complexity of the image texture; the mean represents the overall brightness of the image; the standard deviation represents the roughness of the image texture; dissimilarity represents the diversity of the image texture; uniformity represents the smoothness of the image texture; grayscale inhomogeneity represents the rich changes in the image texture; and the maximum correlation represents the flow direction of the image texture.

[0033] According to a preferred embodiment, the visualization of texture features extracted from the gray level co-occurrence matrix includes:

[0034] Visualize the extracted texture feature statistics to gain an intuitive understanding of image textures and local patterns.

[0035] According to a preferred embodiment, the color features and texture features are combined to identify geographical elements through symbol guides and texture feature maps.

[0036] The method for identifying geographical elements in historical topographic maps based on image features provided by the present invention has the following advantages and beneficial effects: the present invention performs a systematic analysis of color and texture feature extraction, and intuitively demonstrates the effect of geographical element identification through visualization of the results; the present invention combines knowledge from multiple disciplines such as surveying and mapping engineering, computer vision, and image processing, and demonstrates the advantages of interdisciplinary research in solving practical problems; the present invention improves recognition efficiency and accuracy, and can provide practical technical support for historical geographical research. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will describe the technical solutions required in the embodiments or the prior art.

[0039] The accompanying drawings are used for a brief introduction. Obviously, the accompanying drawings described below are some embodiments of the present invention, which are not suitable for ordinary technicians in this field.

[0040] For personnel, other drawings can be obtained based on these drawings without any creative work.

[0041] Figure 1 A flow chart of a method for identifying geographical elements in historical topographic maps based on image features provided by the present invention;

[0042] Figure 2 A schematic diagram of vegetation segmentation by color clustering segmentation provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of water area segmentation for color clustering segmentation provided in an embodiment of the present invention;

[0044] Figure 4 This is a recognition result diagram combining color features and texture features provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the present invention will be described in detail below in combination with the embodiments of the present invention and the accompanying drawings.

[0047] Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0048] The embodiments of the invention and all other embodiments obtained by ordinary technicians in this field without creative work belong to the present invention.

[0049] The scope of protection of the invention.

[0050] This embodiment selects the topographic map of Dadeville, Alabama, U.S. The map is published by the U.S. Geological Survey and was drawn in 1969.

[0051] The present invention provides a method for identifying geographical elements of historical topographic maps based on image features, as shown in FIG1 , the method comprises the following steps:

[0052] Step 1: Preprocessing of historical topographic maps.

[0053] The map image is read using the OpenCV library and converted from the RGB color space to the HSV color space. The HSV color model is based on the Munsell color system, and the design of its three-dimensional coordinate system can ensure the independence of each color component in the model. When one component is adjusted, the other components remain unchanged. The present invention uses the HSV color model to perform color map image segmentation. The color topographic map is converted from the RGB color space to the HSV color space, and the three-dimensional HSV image array is reshaped into a two-dimensional array, where each row represents the three color channel values ​​of a pixel.

[0054] Step 2: Perform color cluster analysis on the map image to obtain images of objects of different colors after cluster segmentation.

[0055] Construct pixel coordinates based on the hue, saturation, and brightness values ​​of the pixel;

[0056] Determine the number of cluster centers based on the color type of the image. In this example, the number of clusters is set to 7;

[0057] Randomly select the initial cluster center in the pixel coordinates;

[0058] Calculate the distance d between each pixel and the cluster center using the following formula:

[0059] d=

[0060] in , They are the coordinates of the cluster center and the current calculated pixel point respectively;

[0061] Assign pixels to the nearest cluster centers to form clusters with the same number of cluster centers;

[0062] Update the position of the cluster center, using the arithmetic mean of the hue, saturation, and brightness values ​​of all pixels in each cluster as the new cluster center until the change in the cluster center is less than 0.0001;

[0063] Assign a label to each cluster to distinguish different clusters;

[0064] Traverse each cluster, for each cluster, create a completely black image of the same size as the original image, set the color of the pixel points belonging to the current cluster to the centroid color of the cluster according to the cluster label, and construct the segmented color image in sections. Figure 2 The schematic diagram of water area segmentation is shown in Figure 3 shown.

[0065] Step 3: Texture feature extraction preprocessing

[0066] Convert the color historical topographic map into a grayscale image, eliminate the redundant color information in the original image, and highlight the core brightness information of the image.

[0067] Step 4: The gray-level co-occurrence matrix extracts the texture features of the image and visualizes the texture features.

[0068] The gray-level co-occurrence matrix extracts statistics that can accurately characterize the texture characteristics of the map by carefully analyzing the spatial distribution and mutual relationship of the grayscale values ​​of pixels in the map. The construction of the gray-level co-occurrence matrix involves two key parameters: distance (d) and direction (θ). The distance parameter defines the spatial distance between the pixel pairs that need to be considered when calculating the gray-level co-occurrence matrix, and the direction parameter defines the relative direction between the pixel pairs when calculating the gray-level co-occurrence matrix. By setting different distance and direction parameter combinations, the gray-level co-occurrence matrix in multiple scales and directions can be obtained, thereby comprehensively and meticulously describing the texture characteristics of the image. In this example, the distance d is set to 1 and the direction θ is set to 0 degrees. The texture features are extracted using the gray-level co-occurrence matrix and the obtained texture features are visualized. Among them, contrast represents the clarity of the image texture; correlation represents the directionality of the image texture; energy represents the roughness of the image texture; entropy represents the complexity of the image texture; the average value represents the overall brightness of the image; the standard deviation represents the roughness of the image texture; dissimilarity represents the diversity of the image texture; uniformity represents the smoothness of the image texture; grayscale inhomogeneity represents the rich changes in the image texture; and the maximum correlation represents the flow direction of the image texture.

[0069] Step 5: Combine color and texture features to identify and classify geographic features in the map.

[0070] In this example, taking contour lines as an example, the contour line segmentation map contains some roads, which can be distinguished based on texture features. Figure 4 As shown in the figure, the visualization of grayscale inhomogeneity and energy shows a clear contrast effect compared with the original image. In the visualization of grayscale inhomogeneity, areas with higher inhomogeneity values ​​are presented in brighter tones, while areas with lower values ​​are presented in darker tones. Similarly, in the energy visualization, areas with higher energy are represented in bright tones, while areas with lower energy are represented in dark tones. The texture complexity of the contour lines is significantly higher than that of the roads, and the difference in brightness between the two is large. The two can be clearly identified based on texture features and color features.

Claims

1. A method for identifying geographical elements of historical topographic maps based on image features, characterized in that: The method comprises the following steps: Historical topographic map preprocessing; Perform color clustering analysis on the preprocessed map image based on color features to obtain images of objects of different colors after cluster segmentation; The gray-level co-occurrence matrix is ​​used to extract the texture features of the image; Visualize texture features; Combine color features and texture features to identify geographical features on historical maps.

2. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: The historical topographic map preprocessing includes: The historical topographic map was converted from RGB color space to HSV color space to enhance the accuracy of color clustering; Reshape the three-dimensional HSV image array into a two-dimensional array for clustering.

3. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: The method for performing color clustering analysis on the preprocessed map image based on color features comprises: Construct pixel coordinates based on the hue, saturation, and brightness values ​​of the pixel; Determine the number of cluster centers based on the image color category; Randomly select the initial cluster center in the pixel coordinates; Calculate the distance d between each pixel and the cluster center using the following formula: d= 。 4. Among them , They are the coordinates of the cluster center and the current calculated pixel point respectively; Assign pixels to the nearest cluster centers to form clusters with the same number of cluster centers; The arithmetic mean of the hue, saturation, and brightness of all pixels in each cluster is used as the new cluster center, and the position of the cluster center is updated until the change in the cluster center is less than 0.0001.

5. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: The images of objects of different colors after segmentation and clustering include: Assign a label to each cluster to distinguish different clusters; Iterate over each cluster and create a segmented color image for each cluster.

6. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: Said The gray level co-occurrence matrix is ​​used to extract the texture features of the image, including: Convert the color historical topographic map into grayscale image to eliminate the redundant color information in the original image; Calculate the gray-level co-occurrence matrix to obtain the texture information of the image; The texture feature values ​​are extracted from the gray-level co-occurrence matrix, where contrast represents the clarity of the image texture; correlation represents the directionality of the image texture; energy represents the roughness of the image texture; entropy represents the complexity of the image texture; the mean represents the overall brightness of the image; the standard deviation represents the roughness of the image texture; dissimilarity represents the diversity of the image texture; uniformity represents the smoothness of the image texture; grayscale inhomogeneity represents the rich changes in the image texture; and the maximum correlation represents the flow direction of the image texture.

7. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: The texture feature visualization extracted from the gray level co-occurrence matrix includes: Visualize the extracted texture feature statistics to gain an intuitive understanding of image textures and local patterns.

8. The method for identifying geographical elements of historical topographic maps based on image features according to claim 1, characterized in that: Said Combining color and texture features, geographic features are identified through symbol guides and texture feature maps.

Images