We present a new segmentation approach by thresholding based on learning strategy. This strategy is based only onto one image and its ideal thresholded version. A decision matrix is generated from each pixel and each gray level. At the moment of new images segmentation, the best solution for each pixel is evaluated by means of K Nearest Neighbors in the decision matrix. Comparative tests were performed on signature, fingerprint and magnetic resonance images.
Citation:
Horacio Andr? Legal-Ayala, Jacques Facon, "Segmentation Approach by Learning: Different Image Applications," iciap, pp.600, 12th International Conference on Image Analysis and Processing (ICIAP'03), 2003