An efficient filtering technique for detecting traffic surveillance in intelligent transportation systems. (17th March 2023)

• Record Type: Journal Article
• Authors: Hemalakshmi, K.; Muthukumaravel, A.
• Is Part Of: International journal of intelligent engineering informatics. Volume 10:Number 6(2022)
• Page Start: 504
• Page End: 521
• Publication Date: 2023-03-17
• Subjects: vehicle detection -- ITS -- intelligent transportation systems -- filtering -- noises -- NLMTF -- noise removal creates the final image -- non-local means and trilateral filter -- Gaussian noise; Artificial intelligence -- Engineering applications -- Periodicals; Engineering -- Computer programs -- Periodicals; Knowledge management -- Periodicals; 620.0028563
• Languages: English
• ISSNs: 1758-8715

Abstract

A prospective system for traffic monitoring, vehicle surveillance, and vehicle control have become unavoidably necessary to support intelligent transportation systems (ITS) due to the rapid increase in vehicle numbers, traffic congestion, and security issues. Due to brightness changes, image subtraction, a standard method for recognising moving objects in films, is inefficient. Optimisation is needed for moving vehicle recognition and tracking despite the many methods available. The best car tracking algorithm for traffic footage with changing lights, backgrounds, and noises uses image and video processing. First, the input video is broken into frames of road visuals in various traffic scenarios. The input film is first split into frames of traffic scenes. A colour model distinguishes between road and car in these photographs. Noise removal creates the final image. Non-local means (NLMs) and trilateral filter (NLMTF), which eliminates Gaussian noise, is suggested here. Experimental results show that the suggested filter outperforms noise reduction algorithms like block matching and 3D filtering (BM3D), Shearlet transforms, and Mean filters in mean square error (MSE), peak signal-to-noise ratio (PSNR), and structure similarity index imatest (SSIM) and Python evaluation.