A Novel Filter For the Detection of Object Boundary and Edges in Noisy and Non-Noisy Images

8 Pages Posted: 10 Mar 2020

See all articles by Sandeep Chand Kumain

Sandeep Chand Kumain

National Institute of Technology, Uttarakhand - Department of Computer Science and Engineering

Kamal Kumar

National Institute of Technology (NIT)

Date Written: February 15, 2020

Abstract

Noise removal, Object detection and Edge identification in an image is an active research problem in Computer Vision, Machine Vision and Digital Image Processing (DIP). Edge detection is considered as one of the basic steps in the DIP for the detection of the object. Edge discovery of the object is one of the big challenges when we have an image with a complex and clutter background or image having a significant amount of noise. Due to the noise in an image their always be a chance of non-true edge generation. The traditional operator like sobel, robert, canny, prewitt, log is able to find the edge pixel but not shown the good result when we focus on a proper boundary and edge detection for the objects in noisy images. In this paper author's proposed a novel filter, which gives the good result in comparison of traditional operators in a qualitative or quantitative parameter. For experimental result analysis, a qualitative and quantitative comparison is done with another well-known operator like sobel, robert, prewitt etc. All experimental analysis is done with the randomly selected images from the MSRA dataset (Salient Object data-set). Noisy images are prepared by mixing the artificial noise using MATLAB with there default value of mean and variance and further these images are used to evaluate the performance of our proposed filter in presence of noise.

Keywords: Sobel Operator, Prewitt Operator, Robert Operator, DIP, Gaussian PSNR

Suggested Citation

Chand Kumain, Sandeep and Kumar, Kamal, A Novel Filter For the Detection of Object Boundary and Edges in Noisy and Non-Noisy Images (February 15, 2020). 5th International Conference on Next Generation Computing Technologies (NGCT-2019), Available at SSRN: https://ssrn.com/abstract=3538586 or http://dx.doi.org/10.2139/ssrn.3538586

Sandeep Chand Kumain (Contact Author)

National Institute of Technology, Uttarakhand - Department of Computer Science and Engineering ( email )

NH 58
Srinagar, Uttarakhand 246174
India

Kamal Kumar

National Institute of Technology (NIT) ( email )

NH 58
Srinagar, Uttarakhand 246174
India

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
54
Abstract Views
379
rank
534,115
PlumX Metrics