A NEW APPROACH OF SEGMENTATION IMAGES BASED ON WAVELET TRANSFORM

Abstract

In this work, we present an approach to region-contour cooperative segmentation that has gained significant interest in recent years. It involves cooperation between region-based segmentation and contour-based segmentation. It exploits the advantages of both segmentation types to achieve a more accurate and faithful segmentation result than what can be obtained using a single technique. Wavelets have been applied to transform the image and reconstruct the model at a higher resolution. Then, the segmentation is performed using Active Contour (Region Scalable Fitting (RSF)). A comparison with the result obtained from active contour segmentation demonstrates the advantages of adopting such cooperation.