A Novel Evaluation Approach for Line Simplification Algorithms towards Vector Map Visualization

Abstract

Abstract:Line simplification is an important method in the context of cartographic generalization, which is helpful for improving the visualization of digital vector maps. The evaluation method for the simplification algorithms is still an open issue when facing applications of vector data, including progressive transmission, web mapping, and so on. This paper proposes a novel evaluation approach for line simplification algorithms based on several factors towards vector map visualization, including the features of displays, map scales, and the ability of the human eye to distinguish pixels. In order to ensure the evaluation of the line simplification algorithms is conducted under the consistent strength of simplification, a measurement approach for the difference between an original line and its simplified one is proposed in this study, and the method of solving the appropriate simplification threshold is presented. With this method, four simplification algorithms are evaluated at five map scales using three evaluation indicators: standard deviation, compression ratio, and simplification time. The experiment and results show the evaluation approach in this study is feasible, and represents a good means in which to facilitate the application of line simplification towards progressive transmission and visualization of vector maps.

Batch Simplification Algorithm for Trajectories over Road Networks

“…Among them is the area of trajectory pre-processing which studies trajectory simplification techniques and algorithms. The trajectory simplification algorithms eliminate some subtraces of the original trajectory [16]; which decreases the data storage space and the data transfer time [17][18][19]. A framework where these areas are observed is proposed in this paper [20].…”

An Optimized Algorithm for Computing the Voronoi Skeleton

“…However, certain simplification strategies do not agree with the simplification requirement derived from Lemma 4. For example, a naive Nth point simplification [38] method merely removes each Nth point from a polygon ignoring its geometry. Another algorithm -Circle simplification [38], aims to group together points forming spatial clusters based on the distance threshold and replace these clusters by a single representative.…”

“…For example, a naive Nth point simplification [38] method merely removes each Nth point from a polygon ignoring its geometry. Another algorithm -Circle simplification [38], aims to group together points forming spatial clusters based on the distance threshold and replace these clusters by a single representative. Li-Openshaw [37] and Rapso [36] algorithms simplify polyline based on spatial pixel (or hexagon-based) grid.…”

Building detection and regularisation using DSM and imagery information

“…The most common data-driven approaches can be classified, in terms of the underlying method, into three groups: (i) the Douglas-Peucker (DP) algorithm (Douglas and Peucker, 1973); (ii) the random sampling consensus (RANSAC) method (Fischler and Bolles, 1981); and (iii) the Hough transform (Hough, 1962). Many of the existing data-driven approaches have utilised an initial solution based on the DP algorithm (Maas and Vosselman, 1999;Wang et al, 2006;Jwa et al, 2008;Sohn et al, 2012;He et al, 2014) because it is easy to implement and is able to maintain the original shape (Song and Miao, 2016). However, the efficiency of the DP algorithm decreases with increasing irregularity of the building boundary points.…”