Image Similarity Detection

The goal of this project is to improve the efficiency of image similarity detection by developing an algorithm called Sc-SIFT that uses sparse coding to detect and extract the features of an image. The image's SIFT feature is derived as a training sample to complete an overcomplete dictionary [1]. It is then used to generate an index. The proposed algorithm is compared to the standard algorithm for image similarity detection. The results indicate that it can improve the detection speed. The image similarity detection algorithm can be used in many applications such as detecting if two images taken from different environments or different angles are similar or not