The Study of Feature Extraction Based on the Optimal Threshold Segmentation Algorithm

Image segmentation technology is crucial for feature extraction. In this paper, the traditional thresholding method has been improved, and the optimal threshold segmentation method is proposed by combining the iterative algorithm and mathematical morphology. Through analyzing the target area of image, the scope of the general gray-scale is determined. Based on the image edge and region, fast iterative algorithm is used to choose the optimum threshold as segmentation criteria. Moreover, morphology open computing is applied to the image threshold segmentation. Then the method is used in processing marine oil spill and solar flares. Different experiments of marine oil spill and solar flares based on the method show that this method can not only extract a clearer definition of interesting areas, but also can show a more clear-cut edge.