Date of Publication :20th March 2018

Abstract: The system proposes new approach in extension with local color and Fast curvelet transform and entropy measurement in RGB Space. Discrete curvelet transform is one of the most powerful approaches in capturing edge curves in an image. The project presents the robust object recognition using texture and directional feature extraction. The system proposes texture descriptors such as Fast Discrete Curvelet Transform (FDCT) based entropy feature which represents better texture and edges and Local Directional Pattern (LDP) which provides textural details about all eight directions. By using these methods, the category recognition system will be developed for application to image retrieval which proves Low computational complexity and high compatibility.The tests are performed more than 12 seat stamp regular scene and shading surface picture databases, for example, Corel-1k, MIT-VisTex, USPTex, Hued Brodatz, et cetera.

Reference :