A Robust Adaptive Multi-Scale Superpixels Segmentation in SEM Image
R. Susithra1, A. Mahalakshmi2, Judith Justin3
1R. Susithra, PG Scholar, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore (Tamil Nadu), India.
2A. Mahalakshmi, Research Scholar, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore (Tamil Nadu), India.
3Dr. Judith Justin, Professor and Head, School of Engineering, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore (Tamil Nadu), India.
Manuscript received on 23 April 2019 | Revised Manuscript received on 02 May 2019 | Manuscript Published on 08 May 2019 | PP: 161-166 | Volume-7 Issue-5S3 February 2019 | Retrieval Number: E11300275S19/19©BEIESP
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: This work presents a region-growing image segmentation approach based on superpixels SEM Image segmentation decomposition. From a first contour- constrained over-segmentation of the input image, the image segmentation into parts is achieved by again and again merging like superpixels SEM image segmentation into parts into fields, regions. This approach raises two key issues: how to compute the similarity between superpixels SEM Image segmentations in order to perform accurate merging and in which order those superpixels SEM Image segmentations must be merged together. In this perspective, we firstly introduce a robust adaptive multi-scale superpixels SEM Image segmentation similarity in which region comparisons are made both at content and common border level. Secondly, we offer a complete merging worked design to small amount of support guide the field, region merging process. Such strategy uses an adaptive merging criterion to ensure that best region aggregations are given highest priorities. This lets to get stretched to a final segmentation into harmony regions with strong division line take as rule. We act experiments on the BSDS500 image dataset to high-light to which extent our method compares favorably against other well-known image segmentation algorithms.
Keywords: Scanning Electron Microscope (SEM), Adaptive Multi Scale Superpixel, Image Segmentation.
Scope of the Article: Image Security