Fast manifold spectral clustering algorithm for intelligent traffic remote sensing image fuzzy edge

2017 3rd International Conference on Applied Materials and Manufacturing Technology

Deng, Cong^1 ; Jia, Zelin^1 ; Li, Shen'An^1 ; Tang, Pengfei^1

School of Natural and Applied Sciences, Northwestern Polytechnical University, Xi'an

710072, China^1

关键词: Absolute coordinate;    Clustering methods;    High resolution remote sensing images;    Intelligent traffics;    Intelligent transportation systems;    Remote sensing images;    Simulation environment;    Spectral clustering algorithms;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/242/1/012107/pdf
DOI  :  10.1088/1757-899X/242/1/012107

来源: IOP

【 摘 要 】

Intelligent transportation system is the future development of traffic systems. Higher precision map requires a higher absolute coordinate accuracy. However, cluster labels cloud be connected to the wrong pixels near the edges of regions, and edges will be misplaced. This paper proposed a fast Manifold Spectrum Clustering Algorithm for the fuzzy-edge of intelligent traffic remote sensing image. we first construct the fuzzy edge of remote sensing image, then analyze and evaluate the influence degree of the fuzzy edge on detailed image, and proposed a spectrum clustering method based on manifold to improve the accuracy and speed to obtain the fuzzy edge of the road in the high resolution remote sensing image. The simulation environment is built by Matlab 2015a, and it is proved that this paper is superior to the existing method in clustering speed and precision performance.