K-Nearest Neighbor Based Enhanced-CBIR System
Somala Ramakishore1, P. Karpagavalli2
1Somala Ramakishore*, Research Scholar, Department of E.C.E, Sri SatyaSai University of Technology & Medical Sciences, Sehore, Bhopal, Indore Road, Madhya Pradesh, India.
Dr. P. Karpagavalli, Research Guide, Department of E.C.E, Sri Satya Sai University of Technology & Medical Sciences, Sehore, Bhopal, Indore Road, Madhya Pradesh, India.
Manuscript received on January 18, 2020. | Revised Manuscript received on January 25 2020. | Manuscript published on February 10, 2020. | PP: 550-553 | Volume-9 Issue-4, February 2020. | Retrieval Number: B7711129219/2020©BEIESP | DOI: 10.35940/ijitee.B7711.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In content based image retrieval is the most widely recognized feature utilized are shape, hues, surface and so on. To improve the exactness of retrieval, it must look on the far side the old style features. The features which could without much of a stretch be extracted from information could be considered. One of such feature is directionality of the picture surface. Directional data can be spoken to in a minimized way by utilizing transform like wavelet, Gabor, Radon and so on. In this proposal, we address this issue of utilizing directional data to build exactness of enhanced-CBIR. Picture retrieval execution is assessed by utilizing Precession and Recall. These calculations are most appropriate for retrieval of textural pictures. Our proposed Enhanced-CBIR system which works combine with KNN algorithm, provides better quality of result compare than the existing CBIR framework.
Keywords: Enhanced, CBIR, KNN, Algorithm.
Scope of the Article: Algorithm Engineering