AI Based Littering Detection and Penalty System

Littering is a threat to the environment as well as public health. Different methods used that aim to resolve such issues are termed as traditional approaches, these approaches include but are not limited to, manual reporting and checking of photographs that have proven to have been ineffective due to delays and inaccuracies. This paper presents an AI automatic system based on sensing, offender recognition and fine enforcement. The system uses surveillance footage and combines it with the following technologies: YOLOv4, DeepSORT, MTCNN&R ArcFace. Activities like throwing away trash are registered offenders are recognized by faces and referenced in a system of database; such offenders are ordered to pay a fine. This automatic method cuts down on manual work, lowers the chance of human mistakes, and helps enforce cleanliness rules in public areas more. The setup does not just lead to cleaner environment by discouraging people from littering, but it also makes the fine process simpler more productive, and more dependable. In the future, we might see improvements like better object spotting more accurate face recognition in tough situations, and links to city systems to make it work on a larger scale and have a bigger effect.