Improved shadow suppression with amalgamation of AGWCD and exemplar based inpainting method

In today’s digital times, the quality of video frames is ubiquitous, and the presence of shadows is undesirable in computer vision applications. Shadow suppression is of paramount significance in crucial application areas, especially in outdoor scene environments. The objects present in the environm...

Full description

Saved in:
Bibliographic Details
Published in:Journal of intelligent & fuzzy systems 2022-01, Vol.43 (3), p.2907
Main Authors: Abin, Deepa, Thepade, Sudeep D
Format: Article
Language:English
Subjects:
Amalgamation
Computer vision
Image quality
Performance evaluation
Performance measurement
Shadows
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In today’s digital times, the quality of video frames is ubiquitous, and the presence of shadows is undesirable in computer vision applications. Shadow suppression is of paramount significance in crucial application areas, especially in outdoor scene environments. The objects present in the environment occlude the light. Most of the work in literature focuses on single shadow regions in a frame or an image. Different methods are proposed in the literature. This challenging area of shadows suppression is addressed with the proposed method, as a novel amalgamation, with Adaptive Gamma Weighted Correction and modified Exemplar based inpainting method. The paper discusses different single shadow scenarios and multiple distributed shadow regions. Across four datasets, and objective evaluation using three performance metrics, the obtained average Entropy of 7.032, ‘Blind Reference Image Spatial Quality Evaluator (BRISQUE)’ of 26.2031, and ‘No-Reference Image Quality Evaluator (NIQE)’ of 3.699 have demonstrated considerable results.
ISSN:1064-1246
1875-8967
DOI:10.3233/JIFS-212823