We present a new segmentation approach by thresholding based on learning strategy. This strategy is based only on one image and its ideal thresholded version. A decision matrix is generated from each pixel and each gray level. At the moment of new image 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.
Segmentation approach by learning: different image applications
2003-01-01
760186 byte
Conference paper
Electronic Resource
English
Segmentation Approach by Learning: Different Image Aplications
British Library Conference Proceedings | 2003
|Image Segmentation by Learning Approach
British Library Conference Proceedings | 2003
|Image segmentation by learning approach
IEEE | 2003
|A Learning Approach for Adaptive Image Segmentation
IEEE | 2006
|A Lattice Approach to Image Segmentation
British Library Online Contents | 2006
|