Machine Learning Based Traffic Anomaly Detection in Video Surveillance
Nowadays, The traffic inconsistency is increasing as the number of motor vehicles on the road are increasing gradually. Various anomalies related to traffic jams, vehicles at zebra crossings, triple seats on bikes, etc. are reported. The manual process of handling violations of such traffic rules is difficult, time-consuming, and requires more manpower. Therefore, an automated traffic monitoring system is proposed so as not to limit existing systems. In this study, two traffic discrepancies for implementation e.g. Vehicles at Zebra Crossing, signal detection, and traffic jams (based on traffic density) are considered. It makes reality visible so that it works much more superior than such systems that depend on finding the metal content of motor vehicles. This present study can be more convenient for catching the lawbreaker and greater for traffic control. This study has been implemented using the Open CV Image Processing Library with Python language. The plan is to carry out the Raspberry Pi hardware platform to construct the system transferable and real-time easy.