A Linear Filtering on Automatic Decomposition and Reconstruction of Dermoscopy Images using Global Thresholding
A. Prabhu Chakkaravarthy1, A. Chandrasekar2

1A. Prabhu Chakkaravarthy, Research Scholar, Sathyabama Institute of Science and Technology, Assistant Professor, St. Joseph’s College of Engineering, Chennai, India.
2Dr. A. Chandrasekar, Department of Computer Science Engineering, St. Joseph’s College of Engineering, Chennai, India.

Manuscript received on 02 August 2019 | Revised Manuscript received on 05 August 2019 | Manuscript published on 30 August 2019 | PP: 4257-4263 | Volume-8 Issue-10, August 2019 | Retrieval Number: J99590881019/19©BEIESP | DOI: 10.35940/ijitee.J9959.0881019
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: One of the fastest growing cancers in human cell is skin cancer. Initially starts from the outer layer of human body and spreads unevenly all over by increasing in diameter. The formation of skin cancer depends on the weakness of skin cell. To find the perfect diagnosis, dermatologist should use the computational method. The identification of skin cancer is not easy on the first stage by the dermatologist but the computational method describes out perfectly. The proposed work describes in extracting the skin lesion using Otsu Thresholding with inverse discrete 2D wavelet transform. The proposed work initiated by pre-processing is to enhance the image, followed by segmentation to extract skin lesion with decomposition of fore ground and background image and terminates with post processing to extract feature like uneven boundary. The performance measure between proposed segmentation and ground truth image, described with accuracy up to 96.69% for ISIC 2016 dataset.
Keywords: Gaussian Filter, 2D Discrete Wavelet Transform, Otsu Threshold, 2D Inverse Wavelet Transform, Canny Edge Detection, Morphological Open, Principal Component Analysis.
Scope of the Article: Component-Based Software Engineering