Reducing Building Conflicts in Map Generalization with an Improved PSO Algorithm

Abstract

Abstract:In map generalization, road symbolization and map scale reduction may create spatial conflicts between roads and neighboring buildings. To resolve these conflicts, cartographers often displace the buildings. However, because such displacement sometimes produces secondary spatial conflicts, it is necessary to solve the spatial conflicts iteratively. In this paper, we apply the immune genetic algorithm (IGA) and improved particle swarm optimization (PSO) to building displacement to solve conflicts. The dual-inheritance framework from the cultural algorithm is adopted in the PSO algorithm to optimize the topologic structure of particles. We generate Pareto optimal displacement solutions using the niche Pareto competition mechanism. The results of experiments comparing IGA and the improved PSO show that the improved PSO outperforms IGA; the improved PSO results in fewer graphic conflicts and smaller movements that better satisfy the movement precision requirements.

A new model combining building block displacement and building block area reduction for resolving spatial conflicts

“…In the first step, the evaluation is performed by evaluating the effectiveness of the model in preserving the buildings’ characteristics in the generalization of buildings. In the second step, the accuracy of the proposed model is evaluated and compared with one of the recent optimization models for resolving spatial conflicts (improved PSO [IPSO], conducted by Huang et al., 2017).…”

“…In addition, the results of comparing the proposed model for resolving spatial conflicts with IPSO indicated that although the search space of the IGA is more compatible with discrete spaces and IPSO works properly in continuous spaces [which the result of study conducted by Huang et al. 2017 confirms], when combining the proposed area reduction model with the IGA with improved objective function in a continuous search space, the IGA shows a better performance than IPSO.…”

“…The reason is that the quality of antibodies in the IGA is simultaneously measured by their objective function values and antibody concentrations. In addition, Huang, Guo, Sun, and Liu (2017) proposed an improved particle swarm optimization (PSO) algorithm for displacement of buildings, which combines PSO with the cultural algorithm computational framework. According to the above, the IGA is adopted as a basis for performing building displacement in the present study.…”

“…However, in congested areas, the displacement alone cannot resolve all the spatial conflicts. So, a combination of the displacement with other generalization operators, such as elimination and aggregation, has been suggested in several studies (Ai & Zhang, 2007; Ai et al., 2015; Bader et al., 2005; Huang et al., 2017; Liu et al., 2014; Ruas, 1998; Sun, Guo, Liu, Lv, et al., 2016; Sun, Guo, Liu, Ma, et al., 2016; Wei, He, et al., 2018; Wilson et al., 2003).…”

A spatial contextual immune genetic algorithm to building cartographic displacement

“…During map generalization, proximity conflicts or symbol overlaps often occur between adjacent buildings or between buildings and other adjacent objects (e.g., roads) because of map scale reduction or symbol exaggeration (Liu et al, 2014). To remove these conflicts, several generalization operators have been employed, such as contextual selection (Wang, Guo, Liu, et al, 2017), aggregation (Ai & Zhang, 2007; Li et al, 2004; Shen et al, 2019), displacement (Ai et al, 2015; Bader, 2001; Bader et al, 2005; Basaraner, 2011; Bobrich, 2001; Burghardt & Meier, 1997; Fei, 2004; Harrie, 1999, 2003; Harrie & Sarjakosk, 2002; Højholt, 2000; Huang et al, 2017; Li et al, 2020; Lichtner, 1979; Liu et al, 2014; Lonergan & Jones, 2001; Mackaness, 1994; Mackaness & Purves, 2001; Peng et al, 1995; Pilehforooshha, et al, 2021; Ruas, 1998; Sahbaz & Basaraner, 2021; Sester, 2005; Sun, Guo, Liu, Lv, et al, 2016; Sun, Guo, Liu, Ma, et al, 2016; Wang, Guo, Wei, et al, 2017; Ware & Jones, 1998; Ware et al, 2002, 2003; Wei, He, et al, 2018; Wilson et al, 2003), and typification (Mao et al, 2012; Regnauld, 2001; Shen et al, 2016; Yan et al, 2021). Among these operators, displacement may be the most frequently used one in a map production environment (Ai et al, 2015; Foerster & Stoter, 2008).…”