Search

The popularity of tag clouds has sparked significant interest in the geographic research community, leading to the development of map-based adaptations known as intrinsic tag maps. However, existing methodologies for tag maps primarily focus on tag layout at specific scales, which may result in large empty areas or close proximity between tags when navigating across multiple scales. This issue arises because initial tag layouts may not ensure an even distribution of tags with varying sizes across the region. To address this problem, we incorporate the negative spatial auto-correlation index into tag maps to assess the uniformity of tag size distribution. Subsequently, we integrate this index into a TIN-based intrinsic tag map layout approach to enhance its ability to support multi-scale visualization. This enhancement involves iteratively filtering out candidate tags and selecting optimal tags that meet the defined index criteria. Experimental findings from two representative areas (the USA and Italy) demonstrate the efficacy of our approach in enhancing multi-scale visualization capabilities, albeit with trade-offs in compactness and time efficiency. Specifically, when retaining the same number of tags in the layout, our approach achieves higher compactness but requires more time. Conversely, when reducing the number of tags in the layout, our approach exhibits reduced time requirements but lower compactness. Furthermore, we discuss the effectiveness of various applied strategies aligned with existing approaches to generate diverse intrinsic tag maps tailored to user preferences. Additional details and resources can be found on our project website: https://github.com/TrentonWei/Multi-scale-TagMap. [ABSTRACT FROM AUTHOR]

AbstractRoad network selection plays a key role in map generalization for creating multi-scale road network maps. Existing methods usually determine road importance based on road geometric and topological features, few evaluate road importance from the perspective of road utilization based on human travel data, ignoring the functional values of roads, which leads to a mismatch between the generated results and people’s needs. This paper develops two functional semantic features (i.e. travel path selection probability and regional attractiveness) to measure the functional importance of roads and proposes an automatic road network selection method based on graph convolutional networks (GCN), which models road network selection as a binary classification. Firstly, we create a dual graph representing the source road network and extract road features including six graphical and two functional semantic features. Then, we develop an extended GCN model with connectivity loss for generating multi-scale road networks and propose a refinement strategy based on the road continuity principle to ensure road topology. Experiments demonstrate the proposed model with functional features improves the quality of selection results, particularly for large and medium scale maps. The proposed method outperforms state-of-the-art methods and provides a meaningful attempt for artificial intelligence models empowering cartography. [ABSTRACT FROM AUTHOR]

Cartographic map generalization involves complex rules, and a full automation has still not been achieved, despite many efforts over the past few decades. Pioneering studies show that some map generalization tasks can be partially automated by deep neural networks (DNNs). However, DNNs are still used as black-box models in previous studies. We argue that integrating explainable AI (XAI) into a DL-based map generalization process can give more insights to develop and refine the DNNs by understanding what cartographic knowledge exactly is learned. Following an XAI framework for an empirical case study, visual analytics and quantitative experiments were applied to explain the importance of input features regarding the prediction of a pre-trained ResU-Net model. This experimental case study finds that the XAI-based visualization results can easily be interpreted by human experts. With the proposed XAI workflow, we further find that the DNN pays more attention to the building boundaries than the interior parts of the buildings. We thus suggest that boundary intersection over union is a better evaluation metric than commonly used intersection over union in qualifying raster-based map generalization results. Overall, this study shows the necessity and feasibility of integrating XAI as part of future DL-based map generalization development frameworks. [ABSTRACT FROM AUTHOR]

Urban villages represent informal residential areas emerging since China's rapid urbanization process. Scientific map generalization of urban villages with scientific maps aids readers in discerning their distribution and making informed decisions concerning them. However, there is still a scarcity of research on the automatic extraction and generalization of urban villages from vector data, which needs to be studied to further improve the expression of maps. To address this problem, this paper presents a methodology for the extraction and generalization of urban villages from Digital Line Graphics. Firstly, a heuristic approach is employed to analyze the atypical morphological characteristics of urban villages. Then, indices based on computational geometry and the modified Hausdorff distance are utilized to quantify these traits. Lastly, an automatic generalization principle for urban villages is offered. The approach was tested in experimental blocks and proved to be effective. It offers a novel method for the automatic extraction and cartography of urban villages. [ABSTRACT FROM AUTHOR]

The partitioning of building groups is a prerequisite for map generalization. Existing methods center on the spatial relationships between individual buildings, neglecting the interrelations among building groups. As such, these methods prove suitable only for identifying building groups at a single scale. In this research, we introduce a novel methodology for partitioning building groups across multiple scales by employing image segmentation, thus enabling the simultaneous recognition of building groups of all scales within a recognition unit. First, image segmentation and merging approaches were employed to extract merged image regions, representing the spatial relationships of building groups. Then, the process commenced with the construction of a graphical representation encompassing buildings and merged regions, wherein the edges were assigned weights indicative of the closeness between groups of buildings. Finally, the graph underwent a progressive segmentation process, and a scale set of building groups was established. Based on the experimental findings, the proposed approach can proficiently partition building groups across various scales. This is demonstrated by the accuracy rate surpassing 90% in the complex experimental area. One can effortlessly derive building groups from the scale set by using a predetermined scale parameter. These findings may serve as a fundamental basis for the implementation of multiscale map generalization and the linkage updating of multiscale spatial data. [ABSTRACT FROM AUTHOR]

Continuous-scale representation of urban blocks is critical for enriching the web-based mapping experience in the modern era. Traditionally, continuous-scale merging methods for surface elements have been employed to deal with landscape features, such as patches and land-cover data. In comparison, the artificially constructed urban blocks offer complex semantics and unique shapes. Traditional methods face challenges in maintaining their fundamental characteristics, which can result in improper merging outcomes. Therefore, this paper proposes a progressive merging method that takes into account the functional semantics and grid features of blocks. Initially, the POI data is used to obtain the functional semantic vectors of the blocks for creating basic information about the starting map. The blocks are then abstracted into a graphical structure whereby a breadth-first algorithm is employed to generate their groups; thereby, forming the target map. The map-based geometric and semantic merging cost formula has been devised to calculate the optimal merging sequence by the A* algorithm. Experiments have been performed by defining the main urban area of Beijing city as the study area. The comparative analysis indicates that the proposed method can obtain more reasonable progressive merging results. [ABSTRACT FROM AUTHOR]

Extracting spatial network structure (SNS) from large-scale origin-destination flow data is an important approach for understanding interregional association patterns and interaction laws. Currently, the extraction of SNS primarily relies on complex network clustering or aggregated statistics with predefined regional constraints. However, these methods often overlook one or more fundamental principles essential for ensuring correctness and accuracy: 1) Aggregation of spatially proximate nodes is necessary when strong interactions exist, whereas separation is preferred in the absence of such interactions. 2) It is crucial to maintain strong interactions between non-spatially proximate nodes. 3) Ultimately, nodes within each group should exhibit spatial continuity. To address these challenges, a multi-hierarchical SNS extraction method is proposed, which focuses on raw node aggregating and generalization, measurement of interaction volume and strength between node groups and strategies for node/edge filtering. The effectiveness and value of the proposed method are demonstrated through a case study using city population migration data. Furthermore, the method provides a general approach for extracting SNSs from any origin-destination flow dataset that includes locations and weights, facilitating effective flow map generalization through aggregation of origin destination (OD) flow data. [ABSTRACT FROM AUTHOR]

Objectives: Curve simplification is of importance in automated map generalization; nevertheless, the Douglas-Peucker (DP) algorithm popularly used in map generalization is not automatic, because a key parameter called distance tolerance (ε) must be given by experienced cartographers and needs to be input before execution of the algorithm. Methods: To solve the problem, this paper proposed a method to automatically calculate ε and by which the automation of the DP algorithm is achieved. The method consists of the following steps: (1) A formula is constructed by the Hausdorff distance for calculating the similarity degree (Ssim) between a curve at a larger scale and its simplified counterpart at a smaller scale. (2) 15 linear rivers are selected, and each of them is manually simplified to get their counterparts at seven different scales. The Ssim of each original river and each of its simplified counterpart at a smaller scale can be obtained using the formula constructed by the Hausdorff distance, and 15×7=105 coordinate pairs consisting of (S,Ssim) can be got, and a function between Ssim and S are constructed by the curve fitting using the coordinates. (3) In the meanwhile, the 15 rivers are simplified using a number of ε, and the Ssim of each original river and each of its simplified counterpart at a smaller scale can be calculated using the formula constructed by the Hausdorff distance. In this way, a number of coordinate pairs (ε,Ssim) are got, and a function between ε and Ssim is constructed by the curve fitting. (4) By the function between Ssim and S and that between ε and Ssim, a formula between ε and S can be deducted. Using the formula ε can be calculated automatically, because in a map generalization task S is usually known. After this step, automation of the DP algorithm is achieved. Results: The experiment results show that (1) The proposed DP algorithm can automatically simplify the rivers in a specific geographical area to get the results at different scales; and (2) the resulting river curves generated by the proposed DP algorithm have a high degree of similarity with the ones made by experienced cartographers. Their average similarity degree is 0.927. Conclusion: The proposed DP algorithm can simplify curve features on maps automatically, and the results are highly intelligent and credible. Although only river data is tested in this paper, the principle of the proposed method can be extended to other linear features on maps. Our future work will be on improving the accuracy of the proposed DP algorithm using more river data so that the algorithm can be used in practical map generalization engineering. [ABSTRACT FROM AUTHOR]

Machine learning is increasingly used as a computing paradigm in cartographic research. In this extended editorial, we provide some background of the papers in the CaGIS special issue Machine Learning in Cartography with a special focus on pattern recognition in maps, cartographic generalization, style transfer, and map labeling. In addition, the paper includes a discussion about map encodings for machine learning applications and the possible need for explicit cartographic knowledge and procedural modeling in cartographic machine learning models. [ABSTRACT FROM AUTHOR]

Map generalization is a complex task that requires a high level of spatial cognition, and deep learning techniques have shown in numerous research fields that they could match or even outplay human cognition when knowledge is implicitly in the data. First experiments that apply deep learning techniques to map generalization tasks thereby adapt models from image processing, creating input data by rasterizing spatial vector data. Because image-based learning has major shortcomings for map generalization, this article investigates possibilities to learn directly from vector data, utilizing vector-based encoding schemes. First, we enhance preprocessing methods to match essential properties of deep learning models – namely regularity and feature description – and evaluate the performance of Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Graph Convolutional Neural Networks (GCNN) in combination with a feature-based encoding scheme. The results show that feature descriptors improve the accuracy of all three neural networks, and that the overall performances of the models are quite similar for both polygon and polyline shape classification tasks. In a second step, we implement an exemplary building generalization workflow based on shape classification and template matching, and discuss the generalization results based on a case study. [ABSTRACT FROM AUTHOR]

The automation of map generalization has been an important research subject for decades but is not fully solved yet. Deep learning techniques are designed for various image generation tasks, so one may think that it would be possible to apply these techniques to cartography and train a holistic model for end-to-end map generalization. On the contrary, we assume that map generalization is a task too complex to be learnt with a unique model. Thus, in this article, we propose to resort to past research on map generalization and to separate map generalization into simpler sub-tasks, each of which can be more easily resolved by a deep neural network. Our main contribution is a workflow of deep models, called DeepMapScaler, which achieves a step-by-step topographic map generalization from detailed topographic data. First, we implement this workflow to generalize topographic maps containing roads, buildings, and rivers at a medium scale (1:50k) from a detailed dataset. The results of each step are quantitatively and visually evaluated. Then the generalized images are compared with the generalization performed using a holistic model for an end-to-end map generalization and a traditional semi-automatic map generalization process. The experiment shows that the workflow approach is more promising than the holistic model, as each sub-task is specialized and fine-tuned accordingly. However, the results still do not reach the quality level of the semi-automatic traditional map generalization process, as some sub-tasks are more complex to handle with neural networks. [ABSTRACT FROM AUTHOR]

Deep learning-backed models have shown their potential of conducting map generalization tasks. However, pioneering studies for raster-based building generalization encountered a common "wabbly-wall effect" that makes the predicted building shapes unrealistic. This effect was identified as a critical challenge in the existing studies. This work proposes a layered data representation model that separately stores a building for generalization and its context buildings in different channels. Incorporating adjustments to training sample generation and prediction tasks, we show how even without using more complex deep learning architectures, the widely used Residual U-Net can already produce straight walls for the generalized buildings and maintains rectangularity and parallelism of the buildings very well for building simplification and aggregation in the scale transition from 1:5,000 to 1:10,000 and 1:5,000 to 1:15,000, respectively. Experiments with visual evaluation and quantitative indicators such as Intersection over Union (IoU), fractality, and roughness index show that using a larger input tensor size is an easy but effective solution to improve prediction. Balancing samples with data augmentation and introducing an attention mechanism to increase network learning capacity can help in certain experiment settings but have obvious tradeoffs. In addition, we find that the defects observed in previous studies may be due to a lack of enough training samples. We thus conclude that the wabbly-wall challenge can be solved, paving the way for further studies of applying raster-based deep learning models on map generalization. Demonstrates the effectiveness of the proposed data structure with multiple evaluation indicators Identifies a "wabbly-wall effect" a challenge in deep-learning backed image based map generalization Proposes a layered data structure that separates a target building and its surrounding buildings to ease the learning task in training deep learning models for raster-based map generalization. [ABSTRACT FROM AUTHOR]

Polyline and building simplification remain challenging in cartography. Most proposed algorithms are geometric-based and rely on specific rules. In this study, we propose a deep learning approach to simplify polylines and buildings based on a graph autoencoder (GAE). The model receives the coordinates of line vertices as inputs and obtains a simplified representation by reconstructing the original inputs with fewer vertices through pooling, in which the graph convolution based on graph Fourier transform is used for the layer-by-layer feature computation. By adjusting the loss functions, constraints such as area and shape preservation and angle-characteristic enhancement are flexibly configured under a unified learning framework. Our results confirmed the applicability of the GAE approach to the multi-scale simplification of land-cover boundaries and contours by adjusting the number of output nodes. Compared with existing Douglas‒Peukcer, Fourier transform, and Delaunay triangulation approaches, the GAE approach was superior in achieving morphological abstraction while producing reasonably low position, area, and shape changes. Furthermore, we applied it to simplify buildings and demonstrated the potential for preserving the diversified characteristics of different types of lines. [ABSTRACT FROM AUTHOR]

For point clusters, the conflict and crowding of map symbols is an inevitable problem during the transition from large to small scales. The cartographic generalization involved in this problem as a spatial decision-making process is usually related to the analysis of spatial context, the choice of abstraction operators, and the judgment of the resulting data quality. The rules summarized by traditional generalization methods usually require manual setting of conditions or thresholds and sometimes encounter special cases that make it difficult to directly match certain rules or integrate different rules together. An alternative method is using a data-driven strategy under AI technology background to simulate cartographer behaviors through typical sample training, such as deep learning. The integration of cartography domain knowledge and deep learning is a better choice to settle generalization decisions. This study uses a combination of domain knowledge and a data-driven approach to introduce graph neural networks into point cluster generalization. First, we construct a virtual graph structure of point clusters using Delaunay triangulation, secondly, we extract spatial features, contextual features, and attributes of each point separately, and then propose a generalization model based on the TAGCN network. Finally, this model is trained with the manually generalized sample to realize the automatic point cluster generalization. The results demonstrate that the proposed model is valid and efficient for point cluster generalization and that this algorithm can better maintain various characteristics of the point cluster in both the local area and the overall map compared to other methods. [ABSTRACT FROM AUTHOR]

The cluster patterns of features in map space represent a comprehensive reflection of individual feature geometric attributes and their spatial adjacency relationships. These patterns also embody spatial cognition results under the Gestalt principle. Describing non-linear spatial cluster patterns as effective regular structures is one of the fundamental tasks in deep learning for recognizing feature cluster patterns. In this study, based on the concept of texture co-occurrence matrices from regular gray-scale images, we utilized Voronoi diagrams to construct the tessellation structure of building polygons. Built upon the foundation of first-order texton co-occurrence matrices, we established three-dimensional texton co-occurrence matrices for building polygons, considered five attributes of building size, shape, orientation, and density, and encompassed 64 different combinations of second-order neighboring directions. This matrix concretizes the latent Gestalt spatial characteristics of building polygon clusters into a three-dimensional sparse matrix. It is then used as an input vector to construct a deep convolutional neural network for recognizing building polygon cluster patterns. Through adjustments and optimizations of neural network structure and strategies, along with validation through practical case studies and comparisons with other models, we have demonstrated the effectiveness of the second-order texton co-occurrence matrix in describing the characteristics of building polygon clusters.

When you zoom in or out of current multi-scale cartographic applications, it is common to feel lost and disoriented for a few seconds because dimensions and map symbols have changed. To make the exploration of these multi-scale maps more fluid, one option is to design maps where the transformations due to scale change are more progressive. This paper proposes to use cartographic generalization techniques to design these multi-scale maps with additional intermediate scales to improve progressiveness. These more progressive maps are tested in a user study with a task requiring multiple zooms in and out. The users perform better with the progressive maps, and in particular, the total quantity of required zooming is reduced compared to maps without additional intermediate scales. However, the survey is not fully conclusive on task performance due to the complexity of such a survey with real maps. This difficulty in assessing how well progressive map generalisation reduces disorientation is discussed and guidelines are proposed to design further studies.

Building outlines are important for emergency response, urban planning, and change analysis and can be quickly extracted from remote sensing images and raster maps using deep learning technology. However, such building outlines often have irregular boundaries, redundant points, inaccurate positions, and unclear turns arising from variations in the image quality, the complexity of the surrounding environment, and the extraction methods used, impeding their direct utility. Therefore, this study proposes a simplification and regularization algorithm for right-angled polygon building outlines with jagged edges. First, the minimum bounding rectangle of the building outlines is established and populated with a square grid based on the smallest visible length principle. Overlay analysis is then applied to the grid and original buildings to extract the turning points of the outlines. Finally, the building orientation is used as a reference axis to sort the turning points and reconstruct the simplified building outlines. Experimentally, the proposed simplification method enhances the morphological characteristics of building outlines, such as parallelism and orthogonality, while considering simplification principles, such as the preservation of the direction, position, area, and shape of the building. The proposed algorithm provides a new simplification and regularization method for right-angled polygon building outlines with jagged edges. [ABSTRACT FROM AUTHOR]

Spatial similarity plays a critical role in the perception and cognition in capturing information from maps; it can be used as a constraint to automate map generalization. Although measuring similarities seems natural to humans, it can be challenging to quantify them. This is especially true when it comes to calculating spatial similarity degrees between groups of spatial objects at varying scales and quantitatively expressing the relations between spatial similarity and change of map scale in multiscale map spaces. Taking road networks as an example, this paper proposes an approach to measuring spatial similarity between a road network at a large scale and its generalized counterpart at a smaller scale. By fitting a power function to three typical types of road networks, this paper provides a formula for expressing the change in spatial similarity as the map scale changes. The proposed quantitative method lays a foundation for using spatial similarity as a constraint during road network generalization. [ABSTRACT FROM AUTHOR]

Building displacement is a common operation to resolve the spatial conflicts between map features, and it has important theoretical value and practical application significance for multi-scale mapping. The prerequisite for a successful displacement operation is that there is extra space around the conflicting buildings into which they can be displaced. Otherwise, additional generalization operators need to be combined to resolve spatial conflicts. Based on this idea, this study proposes a new heuristic spatial conflict resolution framework that mainly resolves the spatial conflicts between buildings and other features in urban villages by combining three cartographic generalization operators: selection, displacement, and aggregation. This method first reduces the density of buildings in the block through selection operation, then resolves the spatial conflicts between buildings and other features through displacement operation, and finally, the aggregation operation is performed to eliminate any remaining conflicts and newly generated conflicts. Experiments were carried out using real urban village data, and visual inspection and quantitative analysis were used to evaluate the experimental results. The evaluation results show that the proposed framework can not only resolve spatial conflicts well, but also maintain the spatial distribution and area balance of the buildings in urban villages. [ABSTRACT FROM AUTHOR]

Computer screens often constrain the level of detail and clarity of displays. High-density data require a predefined strategy to select significant features hierarchically to allow interactive data zooming. Although many methods are available for hierarchically selecting rivers from vector data, some approaches for raster data are better than others for maintaining accuracy when the original river data are in a raster format during generalization. In this study, a raster-based approach is proposed to allow hierarchical superpixel selection in river networks. Linear spectral clustering segmentation was applied to divide the original raster river networks into superpixels at multiple levels. A graph was constructed to organize the generated river network superpixels based on the distances between adjacent superpixels by considering the weights determined by the four types of rules. Finally, the total weight values were ranked, the river-network superpixels were selected according to their weights, and the redundant pixels at the river-network intersections were removed. Compared with the traditional vector selection method, the proposed superpixel river network selection method can effectively consider the characteristics of river width without artificial river grading and preserve the main structure and connectivity features during hierarchical mapping. Notably, the average geometry and density changes decreased by 15.8% and 5.1%, respectively. [ABSTRACT FROM AUTHOR]

Urban villages represent informal residential areas emerging since China’s rapid urbanization process. Scientific map generalization of urban villages with scientific maps aids readers in discerning their distribution and making informed decisions concerning them. However, there is still a scarcity of research on the automatic extraction and generalization of urban villages from vector data, which needs to be studied to further improve the expression of maps. To address this problem, this paper presents a methodology for the extraction and generalization of urban villages from Digital Line Graphics. Firstly, a heuristic approach is employed to analyze the atypical morphological characteristics of urban villages. Then, indices based on computational geometry and the modified Hausdorff distance are utilized to quantify these traits. Lastly, an automatic generalization principle for urban villages is offered. The approach was tested in experimental blocks and proved to be effective. It offers a novel method for the automatic extraction and cartography of urban villages.

A polygonal building aggregation method is proposed in this study to address the shortcomings of existing aggregation methods that mostly focus on maintaining a few features, ignoring the requirement for a clear expression. First, visually conglutinated buildings were clustered considering obstacle elements. Unreasonable triangles, starting from the marginal triangles, were removed to extract the visual conglutinate areas. Bridging modes were designed and local shape features were considered to bridge the visual conglutinate areas. Then, the contours of the aggregated results were established through a minimum loop search on the path connection graph. An experiment was conducted using the data of OpenStreetMap from Nanjing, and the comparative experimental results showed that the proposed method preserved the aggregated results of right-angled and locally shaped features. The advantages of improving the visual clarity of the results were verified by introducing degrees of visual conglutination and complexity.

ABSTRACTPolyline simplification is crucial for cartography and spatial database management. In recent decades, various rule-based algorithms for vector polyline simplification have been proposed. However, most existing algorithms lack parameter self-adaptive capabilities and require repeated manual parameter adjustments when applied to different polylines. While deep-learning-based methods have recently been introduced for raster polyline image simplification, they cannot achieve end-to-end simplification when the input data and output results are vector polylines. This paper proposes a new deep-learning-based method for vector polyline simplification by integrating both the vector and raster features of the polyline. Specifically, a deep separable convolutional residual neural network was first used to extract the convolutional features of each image. Then, the region proposal network was modified to generate proposal boxes using vector coordinates, and these proposal boxes were used to locate the convolutional feature maps of bends. Finally, convolutional feature maps were fed into a binary classification network to identify unimportant vertices that should be omitted for vector polyline simplification. The experimental results indicated that the proposed method can exploit raster and vector features to achieve automatic and effective polyline simplification without prior map generalization knowledge and manual settings of rules and parameters. The polyline simplification results of the proposed method have a higher compression ratio of coordinate points and lower shape deformation and deviation than the results generated by the classic Wang and Müller (WM) algorithm and Support Vector Machine (SVM) based algorithm, which shows the potential of the proposed method for future applications in map generalization.

Shape morphing has been used to generate arbitrary scale maps in the field of continuous generalization in recent years. The morphing approach consists of two main steps: shape characteristic matching and trajectory interpolation. Most of the shape characteristic matching methods are difficult to consider the local and global characteristics of spatial data at the same time, which will result in distortion of interpolation results. This paper proposes a Dynamic Time Warping (DTW) distance-based morphing method for continuous generalization of linear features. In this method, the DTW distance is employed as shape similarity distance to find the optimal correspondence by minimizing the total cost between each pair of vertices, which considers both the local and global structure of two linear features. Firstly, we build a matrix to record the distance between vertices in two linear features at two different scales. Then we use the DTW algorithm to find the optimal warping path to establish the corresponding relationship between the vertices of two linear features. Finally, the linear interpolation method is used to dynamically generate the geometric shape at any intermediate scale. Experiment results demonstrate that the proposed method can generate continuous and smooth geometric shapes with a gradual transformation effect, and can be used for the continuous generalization of linear features. The generalization results are consistent with map representation rules and human cognition.

Simplification of building footprints is an essential task in topographic map generalization from large to medium scales. The traditional rule- or constraint-based algorithms commonly require cartographers to enumerate and formalize as many scenarios as possible. Recently, some studies have introduced deep learning to image map generalization, whose outputs, however, may exhibit deformed boundaries due to pure image input. Vector maps are thus a reasonable solution to avoid such issues because of their accurate, object-based geometric representation. However, few existing studies have aimed to simplify buildings in vector maps with the help of neural networks. Building simplification in vector maps can be regarded as the joint contribution from two elementary operations on vertices of building polygons: remove redundant vertices and move kept vertices. This research proposes a multi-task learning method with graph convolutional neural networks. The proposed method formulates the building simplification problem as a joint task of node removal classification and node movement regression. A multi-task graph convolutional neural network model (MT_GCNN) is developed to learn node removal and movement simultaneously. The model was evaluated with a map from Stuttgart, Germany that contains 8494 buildings generalized from the source scale of 1:5,000 to the target scale of 1:10,000. The experimental results show that the proposed method can generate 80% of the buildings with positional errors of less than 0.2 m , 95% with a shape difference under 0.5, and around 98% with an area difference under 0.1 of IoU, compared to the ground truth target buildings, thus demonstrating the feasibility of the proposed method. The code is available at: https://github.com/chouisgiser/MapGeneralizer. [ABSTRACT FROM AUTHOR]

This paper compiles guidelines for minimum dimensions of cartographic symbology from the literature on map design for printed and digital maps. Such minimum dimensions are relevant not only for the design of cartographic symbology, but also as fundamental parameters affecting the visual hierarchy of the overall map and the cartographic generalization of its elements. However, this review shows that only few studies have actually verified minimum dimensions in controlled empirical experiments, and most authors seem to have based their recommendations on anecdotal evidence or theoretical considerations only. Conclusions for a future research agenda in cartography, ranging from ethical questions to calls for empirical research on minimum dimensions, are drawn from the presented findings. [ABSTRACT FROM AUTHOR]

This paper aims to propose a new approach to calculate the quantitative relations between morphostructural similarity degree and map scale change in multi-scale contour clusters for automatic contour generalization. Terrain lines were extracted by pre-processing of unclosed contour lines, and an indirect quantitative expression method of morphostructural similarity relation was proposed based on terrain line hierarchical trees. Thirteen groups of multi-scale contour clusters with different drainage areas of loess geomorphy were employed to explore the changing regularity of morphostructural similarity indices with map scale. Finally, the quantitative relations between morphostructural similarity degree and map scale change were calculated using 52 groups of points. The results show that power function is the best fit among the candidate functions, and the quantitative relations between the morphostructural similarity degree and map scale change can be expressed using the same power function, which facilitates the automation of contour generalization. [ABSTRACT FROM AUTHOR]

When you zoom in or out of current multi-scale cartographic applications, it is common to feel lost and disoriented for a few seconds because dimensions and map symbols have changed. To make the exploration of these multi-scale maps more fluid, one option is to design maps where the transformations due to scale change are more progressive. This paper proposes to use cartographic generalization techniques to design these multi-scale maps with additional intermediate scales to improve progressiveness. These more progressive maps are tested in a user study with a task requiring multiple zooms in and out. The users perform better with the progressive maps, and in particular, the total quantity of required zooming is reduced compared to maps without additional intermediate scales. However, the survey is not fully conclusive on task performance due to the complexity of such a survey with real maps. This difficulty in assessing how well progressive map generalisation reduces disorientation is discussed and guidelines are proposed to design further studies. [ABSTRACT FROM AUTHOR]

The consistency of geospatial data is of great significance for the application and updating of geographic information in web maps. Due to the multiple data sources and different temporal versions, the tile web maps usually meet the inconsistency question across different layers. This study tries to develop a method to detect this kind of inconsistency utilizing a raster-based scaling approach. Compared with vector-based handling, this method can be directly available for multi-level tile images in a pixel representation form. The proposed cross-layer raster tile map rendering method (CRTMRM) consists of four primary aspects: geographic object separation, consistency rendering rules, data scaling and derivation with super-pixel segmentation, and inconsistency detection. The scale transformation strategy with the super-pixel attempts to obtain a simplified representation. Taking the scale lifespan variation and geometric consistency rules into account, the inconsistency detection of tile maps is conducted between temporal versions, multi-sources, and different scales through actual and derived data overlay analysis. The experiment focuses on features of cross-layer water or vegetation areas with Level 9 to Level 14 in Baidu Maps, Amap, and Google Maps. This method is able to serve as a basis for massive unstructured web map data inconsistency detection and support intelligent web map rendering. [ABSTRACT FROM AUTHOR]

In map multiscale visualization, typification is the process of replacing original objects, such as buildings, using a smaller number of objects while maintaining initial geometrical and distribution characteristics. During the past few decades, many vector-based methods for building typification have been developed, whereas raster-based methods have received less attention. In this paper, a new method for the typification of buildings with different distribution patterns called superpixel building typification (SUBT) is developed based on raster data. Using this method, buildings with different distribution patterns, such as linear, grid and irregular patterns, are first grouped by image connected component detection and superpixel analysis. Then, the new positions for building typification are determined by superpixel resegmentation. Finally, a new representation of the buildings is determined through analysis of the orientation and shape of the buildings in each superpixel. To test the proposed SUBT method, buildings from both cities and countrysides in China are applied to perform typification. The experimental results show that the proposed SUBT method can realize typification for buildings with linear, grid and irregular distributions while effectively maintaining the original distribution characteristics of the buildings.

Point settlement selection is one of the critical tasks in map generalization, which should consider spatial distribution, spatial conflict, proximity to objects, and other factors. By analyzing the current research, spatial structure-preserving and conflict-avoiding methods are proposed in this article for point settlement selection. On the one hand, in view of the shortcomings of the Voronoi diagram in measuring settlement density and spatial distribution, a spatial conflict-avoiding strategy is proposed, which utilizes the minimum distance between adjacent settlements to weight the Voronoi diagram area to avoid local conflicts. On the other hand, in order to solve the problem of excessive temporary solidification of adjacent settlements, an adaptive temporary solidification strategy considering local spatial characteristics is proposed, with an improved novel distance-constrained Voronoi diagram model suggested and employed. Comparative experiments and related quantitative analysis on the National Land Survey of Finland Topographic Database confirm that the above methods have good effect. The conflict-avoiding strategy can effectively prioritize map readability and legibility, while the adaptive temporary solidification method can better deal with the boundary effect of settlement groups.

A polygonal building aggregation method is proposed in this study to address the shortcomings of existing aggregation methods that mostly focus on maintaining a few features, ignoring the requirement for a clear expression. First, visually conglutinated buildings were clustered considering obstacle elements. Unreasonable triangles, starting from the marginal triangles, were removed to extract the visual conglutinate areas. Bridging modes were designed and local shape features were considered to bridge the visual conglutinate areas. Then, the contours of the aggregated results were established through a minimum loop search on the path connection graph. An experiment was conducted using the data of OpenStreetMap from Nanjing, and the comparative experimental results showed that the proposed method preserved the aggregated results of right-angled and locally shaped features. The advantages of improving the visual clarity of the results were verified by introducing degrees of visual conglutination and complexity. [ABSTRACT FROM AUTHOR]

Shape morphing has been used to generate arbitrary scale maps in the field of continuous generalization in recent years. The morphing approach consists of two main steps: shape characteristic matching and trajectory interpolation. Most of the shape characteristic matching methods are difficult to consider the local and global characteristics of spatial data at the same time, which will result in distortion of interpolation results. This paper proposes a Dynamic Time Warping (DTW) distance-based morphing method for continuous generalization of linear features. In this method, the DTW distance is employed as shape similarity distance to find the optimal correspondence by minimizing the total cost between each pair of vertices, which considers both the local and global structure of two linear features. Firstly, we build a matrix to record the distance between vertices in two linear features at two different scales. Then we use the DTW algorithm to find the optimal warping path to establish the corresponding relationship between the vertices of two linear features. Finally, the linear interpolation method is used to dynamically generate the geometric shape at any intermediate scale. Experiment results demonstrate that the proposed method can generate continuous and smooth geometric shapes with a gradual transformation effect, and can be used for the continuous generalization of linear features. The generalization results are consistent with map representation rules and human cognition. [ABSTRACT FROM AUTHOR]

Building outlines are important for emergency response, urban planning, and change analysis and can be quickly extracted from remote sensing images and raster maps using deep learning technology. However, such building outlines often have irregular boundaries, redundant points, inaccurate positions, and unclear turns arising from variations in the image quality, the complexity of the surrounding environment, and the extraction methods used, impeding their direct utility. Therefore, this study proposes a simplification and regularization algorithm for right-angled polygon building outlines with jagged edges. First, the minimum bounding rectangle of the building outlines is established and populated with a square grid based on the smallest visible length principle. Overlay analysis is then applied to the grid and original buildings to extract the turning points of the outlines. Finally, the building orientation is used as a reference axis to sort the turning points and reconstruct the simplified building outlines. Experimentally, the proposed simplification method enhances the morphological characteristics of building outlines, such as parallelism and orthogonality, while considering simplification principles, such as the preservation of the direction, position, area, and shape of the building. The proposed algorithm provides a new simplification and regularization method for right-angled polygon building outlines with jagged edges.

Objectives: In the case of uneven distribution of map features, the density difference breaks the visual balance and reduces the beauty of the map. A large number of spatial features are concentrated in the small area, resulting in the contradiction between the display of features and the size of the area. Map details are not prominent, which increases difficulties in obtaining spatial information and is not conducive to decision analysis. The uneven distribution of feature density in mapping area increases the difficulty of map expression. Conventional map expression cannot solve the problem of visual imbalance.Methods: Based on Gastner-Newman cartogram algorithm, this paper proposes a map transformation algorithm that balances the feature density and achieve consistent public perception. Various cartograms can be generated through five steps: Designing deformation unit scales, configuring density weight combination of features, applying linear diffusion equation, constructing deformation grids and mapping discrete control points. Results: Taking Qingxiu district of Nanning city, China as the research area, we select two scale deformation units of subdistrict and custom grid, and configure density weights with different combinations of point of interest and road network geometry features to generate various cartograms. The deformation measurement values of the deformation map are calculated respectively. After cognitive experiments, 56 questionnaires were ob‐ tained. Regarding the morphological control of the administrative map of Qingxiu District in Nanning, the parameters configurations should be controlled as in Sperimeter>0.975, Sperimeter area ratio>0.961 and Sshape ratio> 0.966. In this way, public cognition can be consistent. Conclusions: This study provides a new idea to solve the problem of uneven density of map features. The deformed map reaches visual balance.The area where the original features are dense increases while the area where the features are sparse decreases. By adjusting the deformation weight parameters to maintain the consistency of public map cognition. Subsequent‐ ly, we will further study how to determine the optimal parameter combination to achieve the optimal defor‐ mation expression and cognitive effect. [ABSTRACT FROM AUTHOR]

Firstly, this paper discusses the change in the way of thinking and working of the cartographers and cartographic engineering through data, big data and space-time big data, which promotes cartography to enter a new era of the “fourth scientific paradigm” characterized by “data-intensive computing”. Secondly, this paper analyzes the functions and features of the digital transformation results of 4D products, and the progress in diversification and individuation of the map products, and the “design - compilation-publication” integrated digital transformation of mapping, and the feature of the representative atlases published in 2021. Finally, by giving two specific examples, i.e., the intellectualization of map generalization and digital mapping, this paper reviews four evolutionary stages of map generalization in the era of artificial intelligence and summarizes the difficulties and breakthroughs in key technologies, and provides a profound conclusion of the research status and the issues to be further explored in the domain. [ABSTRACT FROM AUTHOR]

Road networks are the skeletal elements of topographic maps at different scales, and road selection is a prerequisite for implementing continuous multiscale spatial representations of road networks. The mesh-based approach is a common, advanced and powerful method for road selection in dense road areas in which small meshes are removed and road segments with the least importance in each mesh are eliminated. However, small meshes in a map can be classified into two types: aggregated small meshes and isolated small meshes. The number of the former is small, and that of the latter is large. Existing methods are generally applicable for the latter, and some or even most spatial characteristics will be lost when they are applied to the former; as a result, the road selection quality will be affected. Therefore, as a supplement to the mesh-based selection method, this paper proposed an automatic generalization method of dense road network areas (areas formed by aggregated small meshes) considering spatial structural features as constraints. First, the aggregated areas of small meshes were identified based on the number of adjoining small meshes, and the boundaries of aggregated areas are extracted and used as hard constraints during mesh elimination. Second, the starting meshes were redefined by simultaneously considering the edge features and mesh density of small meshes, and an ordinal elimination algorithm was proposed to eliminate the meshes in the stroke connection direction. Third, road selection was implemented by identifying the starting meshes and sequentially processing the related mesh pairs. This iterative process continued until all mesh densities of the newly formed meshes are beyond the threshold or the problem becomes a simple elimination problem involving two adjoining small meshes or one isolated small mesh. Finally, a 1:10,000 standard topographic road map for Jiangsu Province, China, was used for validation. The experimental results showed that in the aggregated areas with two small meshes, 31% of the areas obtained the same selection results by using the mesh-based method and the proposed method, and the remaining 69% obtained a more compact result with the proposed method. Moreover, for all aggregated areas with more than two small meshes, the spatial distribution structure of small meshes was preserved better by the proposed method. [ABSTRACT FROM AUTHOR]

In the multi-scale representation of maps, a selection operation is usually applied to reduce the number of map elements and improve legibility while maintaining the original distribution characteristics. During the past few decades, many methods for vector building selection have been developed; however, pixel-based methods are relatively lacking. In this paper, a multiple-strategy method for raster building selection is proposed. In this method, to preserve the distribution range, a new homogeneous linear spectral clustering (HLSC) superpixel segmentation method is developed for the relatively homogeneous spatial division of building groups. Then, to preserve the relative distribution density, multi-level spatial division is performed according to the local number of buildings. Finally, to preserve the local geometric, attributive and geographical characteristics, four selection strategies, namely, the minimum centroid distance, minimum boundary distance, maximum area and considering geographical element strategies, are designed to generate selection results. To evaluate the proposed method, dispersed buildings in a suburban area are utilized to perform selection tasks. The experimental results indicate that the proposed method can effectively select dispersed irregular buildings at different levels of detail while maintaining the original distribution range and relative distribution density. In addition, the use of multiple selection strategies considering various geometric, attributive and geographical characteristics provides multiple options for cartography. [ABSTRACT FROM AUTHOR]

Map generalization is one of the core technologies of cartography and multi-scale spatial data transformation. Since the 1960s, the research on the automated generalization of map data has gradually developed and made great progress. Furthermore, there are many intelligent solutions on map generalization. However, due to the limitation of the artificial intelligence technology, these intelligent solutions on map generalization are not really intelligent and practical. In the past 10 years, deep learning, as a presentative artificial intelligence technology, was applied in many fields, and the deep-learning-based researches achieved remarkable results. And thus, many new attempts have been made in the intelligent research of map generalization. Firstly, based on analyzing and abstracting models of the automated map generalization, the necessity of the intelligent research on map generalization is pointed out. Secondly, combining with the development of artificial intelligence, the intelligent map generalization is overviewed. Researches of intelligent map generalization based on traditional machine learning and deep learning are sorted and analyzed, and two common strategies of intelligent map generalization are summarized. Finally, focusing on some hot issues of intelligent map generalization, the development tendency of intelligent map generalization is discussed. [ABSTRACT FROM AUTHOR]

Building displacement is a common operation to resolve the spatial conflicts between map features, and it has important theoretical value and practical application significance for multi-scale mapping. The prerequisite for a successful displacement operation is that there is extra space around the conflicting buildings into which they can be displaced. Otherwise, additional generalization operators need to be combined to resolve spatial conflicts. Based on this idea, this study proposes a new heuristic spatial conflict resolution framework that mainly resolves the spatial conflicts between buildings and other features in urban villages by combining three cartographic generalization operators: selection, displacement, and aggregation. This method first reduces the density of buildings in the block through selection operation, then resolves the spatial conflicts between buildings and other features through displacement operation, and finally, the aggregation operation is performed to eliminate any remaining conflicts and newly generated conflicts. Experiments were carried out using real urban village data, and visual inspection and quantitative analysis were used to evaluate the experimental results. The evaluation results show that the proposed framework can not only resolve spatial conflicts well, but also maintain the spatial distribution and area balance of the buildings in urban villages.

Building pattern extraction is one of the crucial components in map generalization. Previous studies mainly focus on the homogeneity between individual buildings in the collinear patterns, while pay less attentions to the complex collinear patterns, in which buildings are locally heterogeneous. Therefore, we put forward a novel combined collinear pattern to extend the existing pattern typology. Then, we present an axis-matching pattern recognition approach which transforms pattern extraction into an axis-matching problem, taking the advantages of computational geometry and visual perception theories. First, we use the axes of the smallest bounding rectangle to represent buildings. Second, different candidate sets of matched axis pairs are built considering the project distance, length and orientation similarities. Third, the strategy for recognizing combined collinear patterns is designed according to similarity criteria. The proposed approach was tested using real topographic data. The experiment results show that the proposed approach is reasonable and efficient, and the correctness and completeness of extracted combined collinear patterns have been guaranteed. [ABSTRACT FROM AUTHOR]

This paper focuses on learning complex polyline generalization. First, the requirements for sampled images to ensure the effective learning of complex polyline generalization are analysed. To meet these requirements, new methods for segmenting complex polylines and sampling images are proposed. Second, using the proposed segmentation and sampling method, a use case for the learning of complex polyline generalization using the generative adversarial network model, Pix2Pix, is developed. Third, this use case is applied experimentally for the complex generalization of coastline data from a scale of 1:50,000 to 1:250,000. Additionally, contrast experiments are conducted to compare the proposed segmentation and sampling method with object-based and traditional fixed-size methods. Experimental results show that the images generated using the proposed method are superior to the other two methods in the learning and application of complex polyline generalization. The results generalized for the developed use case are globally reasonable and suitably accurate. [ABSTRACT FROM AUTHOR]

The deep learning technology has made a breakthrough in many fields, and the automated map generalization based on the deep neural network is highly expected. Building generalization is a typical generalization case and can be abstracted and interpreted as the process of encoding and decoding. It is worth systematically studying the simulation of building generalization with encoder‑decoder based deep neural networks. Methods: Samples are drawn from vector buildings via the segmentation of spatial data and the conversion between vector data and raster data, which is based on a natural principle of objective generalization (i. e., the legibility principle drawn by scale). The encoder decoder based deep neural networks are trained and tested with these samples to learn buildings generalization and to evaluate this application. Five representative models based on encoder‑decoder structure are constructed for learning of generalizing buildings from 1∶10 000 to 1∶50 000. The experimental data cover 44.1 km×38.6 km and contain 89 539 and 21 732 buildings before and after generalization respectively. Five representative models are EDnet, Unet, ResUnet, Unet++ and Pix2Pix representing the basic multi‑layers encoder‑decoder neural network, the encoder‑decoder neural network with skip connection, the residual encoder‑decoder neural network, the dense encoder‑decoder neural network, and the encoder‑decoder neural network with generative adversarial mechanism respectively. The training process and test results of the three constructed models are compared, and effects of these five models while they are applied to the automated generalization of buildings are analyzed. Results: The pixel accuracies of predicted results during EDnet training and testing are lowest obviously, Unet takes the second lowest place, and the pixel accuracies of results generated by other models (ResUnet, Unet++ and Pix2Pix) are higher and relatively close. Additionally, there are some holes and fuzzy representations in generalized buildings predicted by ResUnet and Unet++, which does not conform with the regular representation of buildings. Generalization results predicted by the Pix2Pix show that most of generalized building edges are clear and generalized buildings can be represented independently and completely. Conclusions: The Pix2Pix is more suitable than the other four models for learning and simulating buildings generalization in our experiment. The encoder decoder based deep neural networks are likely to deduce some knowledge and operators of buildings generalization by learning from samples of the original and generalized data. The encoder‑decoder based deep neural networks have the potential in the learning of map generalization. [ABSTRACT FROM AUTHOR]

This paper aims to propose a new approach to calculate the quantitative relations between morphostructural similarity degree and map scale change in multi-scale contour clusters for automatic contour generalization. Terrain lines were extracted by pre-processing of unclosed contour lines, and an indirect quantitative expression method of morphostructural similarity relation was proposed based on terrain line hierarchical trees. Thirteen groups of multi-scale contour clusters with different drainage areas of loess geomorphy were employed to explore the changing regularity of morphostructural similarity indices with map scale. Finally, the quantitative relations between morphostructural similarity degree and map scale change were calculated using 52 groups of points. The results show that power function is the best fit among the candidate functions, and the quantitative relations between the morphostructural similarity degree and map scale change can be expressed using the same power function, which facilitates the automation of contour generalization.

The consistency of geospatial data is of great significance for the application and updating of geographic information in web maps. Due to the multiple data sources and different temporal versions, the tile web maps usually meet the inconsistency question across different layers. This study tries to develop a method to detect this kind of inconsistency utilizing a raster-based scaling approach. Compared with vector-based handling, this method can be directly available for multi-level tile images in a pixel representation form. The proposed cross-layer raster tile map rendering method (CRTMRM) consists of four primary aspects: geographic object separation, consistency rendering rules, data scaling and derivation with super-pixel segmentation, and inconsistency detection. The scale transformation strategy with the super-pixel attempts to obtain a simplified representation. Taking the scale lifespan variation and geometric consistency rules into account, the inconsistency detection of tile maps is conducted between temporal versions, multi-sources, and different scales through actual and derived data overlay analysis. The experiment focuses on features of cross-layer water or vegetation areas with Level 9 to Level 14 in Baidu Maps, Amap, and Google Maps. This method is able to serve as a basis for massive unstructured web map data inconsistency detection and support intelligent web map rendering.

The network Voronoi diagram has been extensively applied in many fields, such as influence area construction, location selection and urban planning, owing to its high accuracy and validity in space division. Taking advantage of parallel processing and auto-wave division of the pulse coupled neural network (PCNN), an algorithm for generating a weighted network Voronoi diagram is proposed in this paper. First, in order to better accommodate the scenes of urban facility points and road networks, the PCNN is improved. Second, the speed of the auto-wave in the improved PCNN is calculated by the weights of the facility points and the attributes of the related road network. Third, the nodes in the road network are considered as neurons, the facility points are projected onto the nearest road segments and the projected points are treated as initial neurons. The initial neurons generate auto-waves simultaneously, and the auto-waves transmit along the shortest path from neurons to other neurons with the calculated speed until all the neurons are fired. During this procedure, the road network and the corresponding space are assigned to the initial neurons and the weighted network Voronoi diagram is constructed. The experiments on the specific region with the real POIs present the feasibility, applicability and efficiency of the algorithm. [ABSTRACT FROM AUTHOR]

Recognizing building group patterns is fundamental to numerous fields, such as urban landscape evaluation, social analysis, and map generalization. Despite the increasing number of algorithms available for building group pattern recognition, there is still a lack of satisfactory grouping results due to insufficient information and only geometric features being provided to recognition methods. This study aims to provide a novel building grouping method that combines building function and geometric information. We specifically focus on the process of recognizing building groups in topographic maps as a prerequisite to subsequent map generalization. First, the building functions are inferred using the dynamic time warping (DTW) algorithm based on Tencent user density data and POIs (points of interest). Then, two types of constrained Delaunay triangulations (CDTs) are created for each building block, from which several spatial indices, such as the continuity index (SCI), direction, and distance of every two adjacent buildings, are derived. Finally, each building block is modeled as a graph on the grounds of derived matrices and building function information, and a graph segmentation approach is proposed to extract building groups. A case study is conducted to test the proposed approach. The experimental results indicate that the proposed approach can produce satisfactory results, given that the correctness value is above 81.63% for our study area. Comparative studies reveal that the method without building function information is an ineffective grouping method when buildings with different functions are close to each other. In addition, generalization results derived from the proposed method are more in line with those of maps for daily use, as they provide users with more accurate spatial divisions of urban buildings. [ABSTRACT FROM AUTHOR]

Quantitative relations between topological similarity degree and map scale change of multi-scale contour clusters are vital to the automation of map generalization. However, no method has been proposed to calculate the relations. This paper aims at filling the gap by proposing a new approach. It firstly constructed a directed contour tree by pre-processing of unclosed contours, and then developed a quantitative expression of topological relations of contour cluster based on directed contour tree. After this, it employed 108 groups of multi-scale contour clusters with different geomorphological types to explore the changing regularity of topological indices with map scale. Last, it used 416 points to calculate the quantitative relations between topological similarity degree and map scale change by curve fitting method. The results show that the quantitative expression of multi-scale topological indexes is closely related to the contour interval change, and power function is the best fit among the candidate functions. [ABSTRACT FROM AUTHOR]