Robust Content-Based Image Retrieval using ICCV, GLCM, and DWT-MSLBP Descriptors

Abstract

The objective of the Content-Based Image Retrieval (CBIR) system is to retrieve the visually identical images from the database efficiently and effectively. It is a broad research realm with the availability of numerous applications. Performance dependency of CBIR focuses on the extraction, reduction, and selection of the features along with the practice of classification technique. In this work, we have proposed the hybrid approach of two different feature descriptors namely, Global Color Histogram and Multi-Scale Local Binary Pattern (MS-LBP). Furthermore, PCA is used for dimension reduction and LDA for the selection of features. The proposed method is evaluated concerning various benchmark datasets namely Corel-1k, Corel-5k, Corel-10k, and Ghim-10k, and results are compared based on precision and recall values at different thresholds. Euclidean distance and City Block distance are used for classification purposes. The performance study of the proposed work displays it as outperformer than the identified literature methods.