Browse

The Virtual University, Pakistan’s first University based completely on modern Information and Communication Technologies, was established by the Government as a public sector, not-for-profit institution with a clear mission: to provide extremely affordable world class education to aspiring students all over the country.

Using free-to-air satellite television broadcasts and the Internet, the Virtual University allows students to follow its rigorous programs regardless of their physical locations. It thus aims at alleviating the lack of capacity in the existing universities while simultaneously tackling the acute shortage of qualified professors in the country. By identifying the top Professors of the country, regardless of their institutional affiliations, and requesting them to develop and deliver hand-crafted courses, the Virtual University aims at providing the very best courses to not only its own students but also to students of all other universities in the country.

A FRAMEWORK FOR DE-NOISING DIGITAL IMAGES THROUGH MACHINE LEARNING

Download

Author: SHAFAQ NISAR


Citable URI : https://vspace.vu.edu.pk/detail.aspx?id=116

Publisher : Virtual University

Date Issued: 4/20/2018 12:00:00 AM


Abstract

There exist many image processing methodologies while most of them do not provide accuracy when images are corrupted with Random Value Impulse Noise (RVIN). Therefore removal of RVIN from the degraded images is believed as preprocessing step in nearly all of the image processing applications. Identification and removal of RVIN from the digital gray scale images is an operational vicinity of research as the detection of RVIN is hard on various noise ratios. In this research a framework has been proposed which leads to the de-noising of RVIN from the digital gray scale images. The detection of noisy pixels provides significant basis for the de-noising process. This proposed framework consists of two main stages. First stage introduces four statistics based features to illustrate the characteristics of the noisy and noise free pixels in the digital image and the Support Vector Machine classifier used to classify the noisy and noise free pixels. In the second stage, modified version of the LPG-PCA algorithm is used for the de-noising of the noisy pixels while preserving the local image structure. The accuracy of this proposed methodology has been compared with the peak-signal-to-noise ratio (PSNR), a well-known performance measure. Simulation results show that the proposed filters provide better accuracy and performance of de-nosing the random value impulse noise in most of the situations.


URI : https://vspace.vu.edu.pk/details.aspx?id=116

Citation: Nisar, S. (2017), A FRAMEWORK FOR DE-NOISING DIGITAL IMAGES THROUGH MACHINE LEARNING, Virtual University of Pakistan, (Lahore, Pakistan).

Version : Final Version

Terms of Use :

Detailed Terms :

Journal :

Files in this item

Name Size Format
Fall 2017_CS720_ms140400001.pdf 1905kb pdf


Copyright 2016 © Virtual University of Pakistan