An Enhanced K-Means Clustering Based on K-SVD_DWT Algorithm for Image Segmentation
M. Banu Priya1, S. Jayasankari2
1M. Banu priya, Ph. D Research Scholar (Full time), Department of Computer Science, P.K.R Arts College for Women (Autonomous), Affiliated to Bharathiar University, Reaccredited With ‘A’ Grade by NACC, Gobichettipalayam, Tamil Nadu, India.
2Dr. S. Jayasankari, Associate Professor in Computer Science, P.K.R Arts College for Women (Autonomous), Affiliated to Bharathiar University, Reaccredited With ‘A’ Grade by NACC, Gobichettipalayam, Tamil Nadu, India.
Manuscript received on 11 September 2019 | Revised Manuscript received on 20 September 2019 | Manuscript Published on 11 October 2019 | PP: 634-640 | Volume-8 Issue-11S September 2019 | Retrieval Number: K110609811S19/2019©BEIESP | DOI: 10.35940/ijitee.K1106.09811S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© 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: The complete nature of the image geared by various processes but image segmentation plays vital role. For object illustration, image analysis, visualization and image processing task the image is segmented into useful information by image segmentation. The image is segmented with respect to the opted scenario by image segmentation. The image measurements like texture, color and depth are considered by the segmentation. The plant disease can be spotted and classified in the field of agriculture and image segmentation is essential in image processing. Based on the morphological characteristics of plants, the diseases can be classified. Image segmentation is of importance within the field of image process. This work focuses on K-means Singular Value Decomposition (K-SVD) and Discrete Wavelet Transform (DWT) that is associated with Kmeans clustering for effectual image segmentation of leaf. Image segmentation is the basic pre-processing task to segregate the leaves in several image process applications. The most challenge in analyzing the plant images are locating and segmenting plants. Image segmentation accustomed to discover the objects and limits such as lines, curves, etc. in images. K-means cluster algorithmic program is wide employed in image segmentation to its machine simplicity. However, the clump results obtained from K-means heavily rely on the initial parameters. Mostly, these initial parameters are elite through hit and trial rule that ends up in inconsistency within the image segmentation results. In this work, an improved K-means cluster algorithmic program is projected for image segmentation, using a histogram based mostly initial parameter estimation procedure. To boot, the projected algorithmic program needs less user interaction to work out K-means data format parameters. Some experiments are conducted supported numerous gray images to check the projected approach. The experiment results show that the projected approach will improve the K-means based mostly image segmentation results.
Keywords: K-means clustering, Image Segmentation, Leaf disease, Histogram
Scope of the Article: Clustering