Trashbusters: Deep Learning Approach for Litter Detection and Tracking

• DOI: 10.48550/arxiv.2404.07467
• Publication Date: 2024-04-11
• Authors: Kashish Jain; Manthan Juthani; Jash Jain; Anant V. Nimkar
• Keywords: Identification; Tracking (education)

The illegal disposal of trash is a major public health and environmental concern. Disposing in unplanned places poses serious risks. We should try to restrict cans as much possible. This research focuses on automating the penalization litterbugs, addressing persistent problem littering places. Traditional approaches relying manual intervention witness reporting suffer from delays, inaccuracies, anonymity issues. To overcome these challenges, this paper proposes fully automated system that utilizes surveillance cameras advanced computer vision algorithms for litter detection, object tracking, face recognition. accurately identifies tracks individuals engaged activities, attaches their identities through recognition, enables efficient enforcement anti-littering policies. By reducing reliance intervention, minimizing human error, providing prompt identification, proposed offers significant advantages incidents. primary contribution lies implementation system, leveraging technologies enhance operations automate litterbugs.