The liver fibrosis identification based on color 2D wavelet transform for the medical image

Based on wavelet transform possessing of specialties of multiresolution and representation local signal, this paper proposed a multiresolution technique for color medical image representation and segmentation, particularly suited for the noisy and large pathological section images. An orthogonal wavelet transform is initially applied to each color channel of the image, and a multiresolution is built up to a selected scale. In course of processing, the lower frequency image which keeps the segmentation capacity of original image is obtained. Then the Fisher linear discrimination analysis and SVD (singular value decomposition) approach are used to extract fibrosis from other tissues on pathological section images. The experiment results show that the method proposed is effective for the stain and large pathological section images and has some practical value.