Acta Geodaetica et Cartographica Sinica ›› 2020, Vol. 49 ›› Issue (2): 256-266.doi: 10.11947/j.AGCS.2020.20190030

• Cartography and Geoinformation • Previous Articles     Next Articles

Complex linear pattern recognition for regular pond groups

LIU Chengyi, WU Fang, GONG Xianyong, XING Ruixing, LUO Denghan   

  1. Institute of Geospatial Information, Information Engineering University, Zhengzhou 450001, China
  • Received:2019-01-17 Revised:2019-04-19 Published:2020-03-03
  • Supported by:
    The National Natural Science Foundation of China (Nos. 41471386;41801396)

Abstract: The existing polygon linear pattern recognition method is incapable facing to the broken area in the regular pond group. In order to solve the problem, a complex linear pattern recognition method for the regular pond group is proposed. Firstly, the spatial characteristics and cognitive characteristics of the regular pond group are analyzed, and the multi-level cognitive order, which is from major-and-minor relationship, parallel relationship, linear pattern to complex linear pattern, is proposed. Secondly, the identification methods of major-and-minor relationship, parallel relationship, and linear pattern are designed. Finally, the identification model of complex linear pattern pond group is constructed. Experiments show that the proposed method can eliminate the adverse effects of broken area on the linear pattern extraction of regular pond groups, and effectively improve the quality of complex linear pattern recognition.

Key words: cartographic generalization, pattern recognition, multi-level cognition, major-and-minor relationship, complex linear pattern

CLC Number: