An Effective Content Based image Retrieval Utilizing Color Features and Tamura Features
El Mehdi, Hassan Silkan · 2019
In In today's real life applications complexity of multimedia contents is significantly increased. This is highly demanding the development of effective retrieval systems to satisfy human desires. In practice, appropriate computer vision and image processing techniques are usually employed to obtain image visual features. Central to functional Content Based Image Retrieval (CBIR) system is effective indexing and fast searching of images based on the visual features. Effective indexing is also essential to make CBIR system scalable for large image databases and incorporation of advanced technique. Usually, CBIR system cannot be developed with only one type of visual features. As such, a combination of different visual features at different levels is needed to identify and classify images in different contexts. In this paper, we propose CBIR system based on color and texture features the color features are represented by Color moments and HSV color histogram and texture feature is represented by Tamura. Detailed experimental analysis is carried out using precision and recall for the two datasets: colored Brodatz textures and KTH TIPS. The Analysis is also performed to compare of the proposed method with the existing similar best methods