Paper Title
A Smart Accident Detection and Recovery Notification Solution

Accident discovery grounded on computer vision through CCTV surveillance has come a healthy but difficult task. In this project, discovery of road accidents is proposed. The proposed frame uses axis constraints for accurate object discovery, followed by an effective, centralized, and grounded GMM(Gaussian Mixture Model) algorithm to track material. The proposed frame provides an effective system to achieve a high discovery rate and a low false alarm rate in common business CCTV surveillance footage. This frame was estimated using the proposed data set under colorful conditions, such as bright daylight, low visibility, rain, and snow. This frame has been effectively designed and opens the way for the development of vehicle accident discovery algorithms in real time. This model also uses a geopy library to record the real-time location, and we can send an alert message to the near police station and hospitals using the accident image. So when they see the image, they get the necessary resources needed, and recovery is really easy in lower time. Along with this an alarm buzzer is made to notify people. Keywords - GMM Algorithm, Accident Detection, CCTV Surveillance, Dataset, Geopy.