›› 2006, Vol. 8 ›› Issue (1): 103-109.

• ARTICLES • Previous Articles     Next Articles

Research on High Resolution Remote Sensing Image Segmentation Methods Based on Features and Evaluation of Algorithms

MING Dongping1, LUO Jiancheng1, ZHOU Chenghu1, WANG Jing2   

  1. 1. The State Key Lab of Resources and Environment Information System, IGSNRR, CAS, Beijing 100101, China;
    2. School of Computer, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2004-06-18 Revised:2005-06-14 Online:2006-03-25 Published:2006-03-25

Abstract: Image segmentation is a key technique in image processing and computer vision field. From the point of view of geo-processing and application of remote sensing images, this paper emphasizes the importance of image segmentation for information extraction and targets recognition from remote sensing images and sets a classification system of common remote sensing image segmentation methods. In addition, this paper states the thoughts of high resolution RS image segmentation methods evaluation and tests it by evaluating four typical image segmentation algorithms based on features with six images qualitatively and quantitatively. The four typical image segmentation algorithms are Max-Entropy (ME), Split&Merge (SM), improved Gauss Markov Random Field(GMRF) and Orientation&Phase(OP). In the qualitative evaluation, this paper analyses these algorithms in terms of their rationale and gets a rough evaluation. In the quantitative evaluation, image complexity is taken into account firstly and five measures are employed. The five measures are removed region rumber, non uniformity within region measure, contrast across region measure, variance contrast across region measure and edge gradient measure. The qualitatively and quantitatively evaluation results are important to perform the optimal selection of segmentation algorithm in practical work. In the end, this paper draws some conclusions about high resolution remote sensing image segmentation and enumerates the flaws of image segmentation methods evaluation, especially it concludes the application prospect of high resolution RS image segmentation.

Key words: high resolution remote sensing, image segmentation, feature, information extraction, evaluation of algorithms