Shearlet Transform Approach to Medical Image Enhancement

Enhancement of low contrast medical images is important for diagnosis of diseases. This paper presents a new approach for image enhancement using Shearlet Transform (ST). Enhancement includes increasing the contrast of the image and edge enhancement. The algorithm is based on modification of Shearlet coefficients by point-wise processing and reconstructing the modified coefficients. For further improvement of the processed image, the enhanced image is followed by adaptive histogram equalization. ST can fully capture the directional and other geometrical features at various scales and directions. In addition, it is highly effective in detecting both location and orientation of the edges, hence enhancement using ST provide good results. Ultrasound, mammogram and retinal images were examined. The proposed ST is compared with Contourlet and Wavelet Transform in terms of Entropy and is observed that the proposed method shows better enhancement of images.