LOGO MATCHING AND RECOGNITION IN VIDEO CLIP

We present a method for detecting the logos from videos. We first include a basic feature-matching algorithm using SIFT, nearest-neighbor matching and RANSAC. Large numbers of features can be extracted from typical images .The features are unique, which allows a single feature to be correctly matched with large database of features, providing a base for object and scene recognition. . Video can easily be divided into frames. These individual frames can be used as input images to check against reference logos from archive. As output count of all logos appearing in the video can be displayed.