Segmentation of high resolution remote sensing images based on improved JSEG algorithm
Tao Fang · Journal of Xi'an University of Science and Technology · 2007
More reasonable segmentations on high resolution remote sensing images are generated from an improved JESG algorithm in terms of the characteristics of rich detail information and clear geometric structure.Our algorithm employs an incremental growing approach to accomplish initial segmentation.And an integrated criterion including the color and shape information is used to merge the over-segmented regions.Experimental results demonstrate that the improved algorithm gains better performance because of the complete use of the metrics of high resolution remote sensing images.