SISI Metric: Image Quality Assessment from Edge Information based on Local Polynomial Approximation Model
1K.Rajkumar, Assistant Professor,Achariya College of Engineering Technology,Achariyapuram,Villianur,Puducherry, India.
2Dr.V.Alamelumangai, Professor,Department of Electronics and Instrumentation,Annamalai University,Annamalai Nagar,Chidambaram, India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 3044-3050 | Volume-8 Issue-8, June 2019 | Retrieval Number: H7059068819/19©BEIESP
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 quality assessment of an image plays an important in image processing application systems. Smoothness is one of the most determining factors in the perceptual assessment of image quality. In order to easily and rapidly forecast the image quality, Smoothness and Sharpness computation plays a vital role in the optimal design image processing algorithms. This paper introduces a novel image quality index, Sensed Image Smoothness Index (SISI) that defines the image quality irrespective of the noise. The high quality images are acquired from the public repository, Computation Visual Cognition laboratory that composes of natural scenery images. The acquired images are added by three different noises, namely gaussian, poisson and Heteroscadiscity at different levels and then removed by Local Polynomial Approximation (LPA) model that define the local and global features of an image. With the global features, the SISI metric is designed under multiple thresholds. Initially, the edge oriented information is retrieved from the denoised image. Relied upon the weighting coefficient, the developed SISI metric depicts the accurate information for achieving better image quality. It is evident from the results that the proposed model achieves better explicability of an image.
Keyword: Quality assessment, Smoothing, Natural scenery images, Local Polynomial Approximation and Multiple thresholds.
Scope of the Article: Image Processing and Pattern Recognition.