Wu, Lan, Sun, Cheng, Shi, Chen and Zen – AN IMPROVED A.N.N.-BASED LABEL PLACEMENT METHOD CONSIDERING SURROUNDING FEATURES FOR SCHEMATIC METRO MAPS 2024

Description

Published on August 19, 2024, in the esteemed ISPRS International Journal of Geo-Information, this 21-page open-access article introduces a cutting-edge ANN-based method for enhancing label placement on schematic metro maps. Authored by a team of seven researchers from Southwest Jiaotong University, this work significantly improves upon prior ANN techniques by integrating the crucial influence of surrounding map features and established cartographic intelligence. The paper is rich with visual aids, including 24 figures showcasing metro maps, technical diagrams, and comparative results. Quantitatively, it achieves remarkably low label overlaps (e.g., 0.00% label-point overlap compared to 4.17% in benchmarks) and enjoys a high user preference of approximately 75% over existing methods. This well-funded research offers a highly trustworthy and practical advancement for automating the generation of high-quality, user-friendly schematic metro maps.