With the high percentage of vehicle ownership in the India, parking has become a conflicting and confusing situation for the people. Cars are parked everywhere in a reckless manner around us which has become the root cause of all the major traffic jams. With the growing number of vehicles, this problem of illegal parking is increasing day by day. Even if the cars are parked illegally there are no measurements taken in order to curb it. Not much people are fined and therefore keeps on parking illegally. People don’t know about the rules that whether to park a car at a particular location or not and in turn park anywhere they like causing traffic jams and even accidents in some cases.
Current solution that is present to illegal parking detection is very limited and require videos or live videos of the surveillance then this video looks for the hardcoded places in the image/frame to look if a vehicle is currently present in that place or not and if it is then it looks for the time it remains there .After a certain period of time it gives the illegally parked detection sign to the motor. The detected place is then send to the official which checks the video to confirm if the vehicle is parked illegally or not.
Using AI based approach to detect illegal parking of vehicles (Cars) from an image. The model will receive an image of parked car through the user of the app , then it will try to predict the status of parking based on the background details (Traffic Signs , Crowded (market) area, Open field, Traffic , On road , number of side cars etc. ). It will do the prediction for each car that is present in the image.
Download the trained model from https://drive.google.com/open?id=1YIHtkgSI_7-zth_ymBditxFLu8X-2dxF
for more details refer to ProjectReport.pdf