Improved image segmentation method based on morphological reconstruction

摘要

Image segmentation is a key step in linking up of image processing and image analysis and a big problem in computer vision. Morphological reconstruction is an important method of image edge segmentation. Watershed transformation is a widely used image segmentation tool base on morphological reconstruction. The traditional watershed transformation method was poor in divide objects in different size because single threshold value can't eliminate noises in large object on top and enhance the edges of small object simultaneously. This paper proposed an improved image segmentation method based on morphological reconstruction. Using erosion operation, dilation operation, catchment basins for all size of objects were morphologically marked and re-shaped, which ensure watershed transformation in the final step segmentation the image accurately. The Experiment results showed the proposed method is more accurate in segmentation of complex images than traditional methods.