Optimization Technique for Image Mosaicing using Local Visual Descriptor

Since last few decades, in real time applications image mosaicing has been a challenging domain for image processing experts. In computer vision, Image mosaicing is one of the most important domain of research. The image mosaicing can be done using two different techniques. The first is the direct method and the second one is feature based method. Image Mosaicing technique is basically done into 5 phases, which includes; feature extraction, registration, stitching, warping and blending. It has vast utilizations in the field of 3D image reconstruction, video conferencing, satellite imaging and several medical as well as computer vision fields. This paper presents the review of feature detection techniques for image mosaicing using image fusion. Initially, the input images are stitched together using the popular stitching algorithms i.e. Scale Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF). To extract the best features from the stitching results, the blending process is executed by means of Discrete Wavelet Transform (DWT) using the maximum selection rule for both approximate as well as detail-components.