A Comprehensive Analysis of Feature Extraction and Retrieval Techniques used in Content-Based Image Retrieval Systems

In this paper, the literature review about various feature extraction methods adopted for extracting feature of query image and database images has been discussed. There are main two approaches to implement content based retrieval system. First is the conventional machine learning methods another is deep learning convolution neural network architectures. Efforts are made for detailed survey of both the machine learning and deep learning approaches for the purpose of extraction of most important salient features of images directly affecting the retrieval performance for classifying the images in particular category and finally retrieving most relevant top images.