Multiresolution wavelet decomposition and neuro-fuzzy clustering for segmentation of radiographic images

Segmentation of medical images is a challenging problem in the field of image analysis. Several diagnostics are based on proper segmentation of the digitized image. Segmentation of medical images is needed for applications involving estimation of the boundary of an object, classification of tissue abnormalities, shape analysis, contour detection and texture segmentation. Despite the existence of several techniques, segmentation of specific medical images still remains a crucial problem due to the complex nature of most medical images. A multiresolution image representation approach is used for better analyzing the information present in an image. In our work, we use multiresolution wavelet decomposition to reconstruct the original image such that it contains all the salient features relevant to segmentation and is devoid of the low frequency noise and texture information that can be ignored while segmenting the image. An unsupervised neural network with fuzzy learning rules is then used to segment the reconstructed image.