Pipe belt conveyors provide sustainable solutions for environmentally sensitive or topographically complex powdered and bulk-solid handling processes; belt rotation is one of the most common faults of pipe belt conveyor, and the current pipe belt conveyor project commonly adopts manual periodic inspection method to detect faults, which has the disadvantages of high danger, low efficiency, high labor intensity of inspection staff, and untimely fault discovery leading to serious engineering accidents. In view of the above problems, this paper investigates the research of conveyor belt torsion detection method of the pipe belt conveyor based on image processing, analyzes the mechanism of conveyor belt rotation determination, puts forward an edge detection method based on OUST-Canny operator, and extracts straight line features of conveyor belt edges by using Hough transform and analyzes the processing, and ultimately realizes the detection of belt torsion faults of the pipe belt conveyor. This method greatly simplifies the process of linear feature extraction in complex environments and helps to realize the intelligentization of the pipe belt conveyor, which ensures the safe and efficient transportation of the pipe belt conveyor and at the same time realizes the green, energy-saving and sustainable development.