Acta Geodaetica et Cartographica Sinica ›› 2015, Vol. 44 ›› Issue (5): 526-532.doi: 10.11947/j.AGCS.2015.20130497

Previous Articles     Next Articles

Optimization Approach for Multi-scale Segmentation of Remotely Sensed Imagery under k-means Clustering Guidance

WANG Huixian1,2, JIN Huijia1, WANG Jiaolong3, JIANG Wanshou1   

  1. 1. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China;
    2. Key Laboratory of Technology in Geo-spatial Information Processing and Application System, Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;
    3. Hebei Provincial Institute of Cartography, Shijiazhuang 050031, China
  • Received:2013-12-26 Revised:2014-11-04 Online:2015-05-20 Published:2015-05-27
  • Supported by:

    The National Basic Research Program of China(973 Program)(No. 2011CB707105);The National High-tech Research and Development Program of China (863 Program) (No. 2013AA12A301);Program for Changjiang Scholars and Innovative Research Team in University(No. IRT1278)

Abstract:

In order to adapt different scale land cover segmentation, an optimized approach under the guidance of k-means clustering for multi-scale segmentation is proposed. At first, small scale segmentation and k-means clustering are used to process the original images; then the result of k-means clustering is used to guide objects merging procedure, in which Otsu threshold method is used to automatically select the impact factor of k-means clustering; finally we obtain the segmentation results which are applicable to different scale objects. FNEA method is taken for an example and segmentation experiments are done using a simulated image and a real remote sensing image from GeoEye-1 satellite, qualitative and quantitative evaluation demonstrates that the proposed method can obtain high quality segmentation results.

Key words: multi-scale segmentation, k-means clustering, guidance optimization, FNEA, Otsu threshold method

CLC Number: