Abstract:
Outliers, which were previously viewed as noisy information in statistics, have become a significant issue under extensive study across various academic disciplines and professional domains. In this study, we examined how urban traffic assessments make use of outlier identification techniques. Outlier detection wastes a significant portion of the workforce on road traffic analysis, which negatively affects the entire economy around the world. There is an abundance of literature and research available for the investigation of traffic jams and their effects. Unfortunately, the outcome wasn't quite fulfilling. The primary purpose of city traffic monitoring is to spot anomalies in traffic statistics. Despite technological advancements in contemporary research, the task of detecting abnormal events in surveillance video systems remains difficult and necessitates extensive human intervention. Anomaly detection poses major problems within the realm of computer vision. According to computer vision methodologies, the field of autonomous evaluation of footage from metropolitan surveillance cameras is rapidly expanding. For a while, highways have effectively detected and classified vehicles using classical visual surveillance techniques like motion tracking and background estimation. This framework provides a better understanding of the intuition, constraints, and advantages of current outlier urban traffic detection systems. Consequently, professionals can get some direction on choosing the best approaches for their particular scenario.
