Characterization of Aluminium Oxide Nanoporous Images using different Segmentation Techniques
Parashuram Bannigidad1, Jalaja Udoshi2, C. C. Vidyasagar3
1Parashuram Bannigidad, Dept. of Computer Science, Rani Channamma University, Belgaum, India.
2Jalaja Udoshi*, Dept. of Computer Science, Rani Channamma University, Belgaum, India.
3 C. C. Vidyasagar, Dept. of Chemistry, Rani Channamma University, Belgaum, Karnataka, India.
Manuscript received on September 16, 2019. | Revised Manuscript received on 24 September, 2019. | Manuscript published on October 10, 2019. | PP: 2491-2497 | Volume-8 Issue-12, October 2019. | Retrieval Number: L34311081219/2019©BEIESP | DOI: 10.35940/ijitee.L3431.1081219
Open Access | Ethics and 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 characterizations of the nanoporous membrane require appropriate segmentation of the pores. The segmentation technique used for one category or class of image may not be suitably applied to the other class of images. Selecting the ubiquitous segmentation technique that can be applied to all image types is a challenging issue. In spite of several decades of research, there is no ubiquitously accepted method for image segmentation and therefore it remains a challenge in digital image processing. The objective of the present study is to automate the system with five different segmentation techniques; global thresholding, active contour, K-means, region growing and watershed to extort the pore from the experimental Al2O3 FESEM images, synthesized using different anodizing parameters. The geometrical features; nanopore wall thickness, pore size and porosity are computed for all the experimental images with the proposed segmentation techniques and are compared with the manual results. The variation in the proposed and manual results, referred to as error is computed. It is observed that the average error for the segmented pore characteristics using global thresholding, active contour, K-means, region growing and watershed are 7%, 14%, 13%, 18% and 20% respectively. The analysis predicts, error in using global thresholding segmentation technique is least, as compared to other methods and thus, it is considered the most appropriate segmentation technique among the five methods mentioned in the present study to segment the experimental Al2O3 FESEM images.
Keywords: Active Contour, Global Thresholding K-means, Region Growing, Watershed
Scope of the Article: Image analysis and Processing