Similar Image Retrieval Based on Grey-Level Co-Occurrence Matrix and Hu Invariants Moments Using Parallel Computing

摘要

In the previous years, several researchers have presented various techniques and also various algorithms for a correct and a dependable image retrieval system. This paper goal is to build up an image retrieval system that retrieves the most similar images to the query image. In this method, the Hu invariants moments and the grey-level Co-occurrence Matrix (GLCM) features extraction methods are performed. Furthermore, with the purpose of boosting up the system performance, multi-thread technique is applied. Later, Euclidian distance measure is performed to compute the resemblance between the query image features and the database stored features. And as shown from the results, the execution time has been minimized to 50% of the conventional time of applying both algorithms without multi-thread. The proposed system is evaluated according to the measures that are used in detection, description and matching fields which are precision, recall, accuracy, MSE and SSIM measures.