Analysing Traffic Using a 360-degree camera and ThingSpeak Analytics: a prototype

Road work planning, vehicle accident hotspot, and congestion observation are several developments achieved from traffic monitoring results. The constant increase in traffic and population raises the reliance on surveillance systems to provide accurate results, which would lead to road safety improvements. Compared to standard surveillance cameras, our implementation of a 360 camera captures the maximum field of view, essentially increasing efficiency by reducing the number of required cameras. This study proposes two solutions for detection, analysis, and data visualisation of traffic. The initial method identifies contours and verifies them by grading their size and location, whereas the second method utilises deep learning algorithms, detecting and calculating the probability of each object in the video stream. The project is constrained due to the following factors, as shown in this video: The project was not tested on the street due to ethical concerns; thus, the image processing model could not be deployed