Abstract: Data mining is the process of extracting the needed information and knowledge from the database based on the user’s expectance. In Image mining, Similarity search is one of the ongoing research fields for efficient image retrieval. For similarity search, many algorithms based on hashing and genetic were implemented. The aim of these algorithms is to retrieve the relevant images which are similar to the given query image based on feature extraction method. It estimates the fitness value for binary bit generation and will obtain the optimal solution for image retrieval. Among the reviewed papers four datasets were commonly used. The datasets used are MIRFLICKR, CIFAR-10, NUS-WIDE and SIFT-1M which provides clarity for image and specificity for querying. Mostly the implementation was conducted in Matlab which uses Fortan for version 1 and C for commercial use.

Keywords: Similarity search, Hashing, Genetic algorithm, Feature extraction method.