Real-Time Truck Crane Detection and Tracking for Protecting Transmission Lines

Abstract

One of the major threats to the Extra High Voltage (EHV) transmission lines is the external force or intrusion incurred by construction trucks especially a truck crane. Several accidents occurred when the truck crane drove into the area under transmission lines and stretch its arm (jib) which broke the lines. In this paper, a video surveillance system is proposed for detecting and tracking truck cranes. By using the codebook algorithm, the foreground objects are detected and then cranes can be recognized by using a Gaussian mixture model. A Kalman filter based tracking algorithm is invoked to monitor the movement of the crane. The jib of the crane is then detected by using a set of combined criterions as well as its extension angle is computed by using Hough transform. If the crane is parked over a certain period of time, or the jib extension angle exceeds predefined thresholds, the remote acousto-optic alarm system will be activated to deliver an alarm signal to the operators of truck crane. The preliminary experiments showed that the system is able to achieve automatic detection of truck cranes.