International Journal of Computational Intelligence Research

This paper presents a novel approach in digital image searching and matching. The given key image is converted into gray scale image and after that a matrix is computed with gray scale values of the key image. Then we are collecting the diagonal key elements for diagonal searching key sequence. Using Pair wise sequence alignment we are trying to match the key with available images in the large data base. Initially we discussed various techniques used in digital image searching and matching in this paper. This new algorithm 2DTO1D is a general algorithm for all kinds of image searching and matching. The applications of digital image matching includes medicine, microscopy, remote sensing, astronomy, defense, materials science, manufacturing, security, robotics, document processing and metallography. Face recognition is very essential in the field of criminology. Image searching and matching are very difficult task in image processing, there are several algorithms for image matching. But still needs more optimization for image matching. Image matching algorithms are used to extract images from a very large database and modified large images. Using this we can match criminal photo from a large database. Image processing plays vital roles in satellite images, medical images, geographical images, remote images and digital source images. Image recognition, feature extraction and pattern matching needs improvements in Image processing. There are several methods for image searching and matching, but we need new optimized technique for image searching and matching. This new 2DTO1D approach is tried to give optimized solution in digital image matching.