Roads in tunnels vary in many ways from certain sections of open roads. For many drivers, the subway is an unusual driving area in a road network that can create even pressure. In conjunction with the Advanced Learning Network known as the Object Detection and Monitoring System, a quick and standard tracking system used for detecting automatically and unexpected incidents monitors on CV tunnels, Writing weight acquisition, (2) standings, (3) people outside the tunnel (4) fire. Odts accepts the video frames as input to obtain the location binding (BBox) result for object binding and compares BBoxes of current and previous video images to give each moving object a unique identification number. With this system, you can track a moving object over time, which is unfamiliar with the normal object detection settings. An in-depth learning model was developed in the ODTS with a series of graphically designed (AP) data sets of 0.8479, 0.7161, and 0.9085 values for direct, automatic, human, and product items, respectively. Subsequently, based on the in-depth learning model developed, the ODTS video risk assessment system was evaluated using four risk video recordings for each risk. This allows the system to detect all crashes in 10 sec. The ODTS acquisition capabilities can be automatically upgraded without changes to program codes if the training database is improved.