Target Feature Extraction Technique Based on Stacker Vision System

Under the huge pressure of the rapid increase of container terminal throughput, the automation and intelligence of the container stacker is the key to ensure the stable and efficient operation of intelligent ports. Traditional ports generally control the stacker height machine manually for container transportation, which has the problems of high labor intensity and low operation efficiency. To address this problem, this paper proposes an image processing-based target recognition and positioning detection method to precisely adjust the position of the stacker spreader through the offset amount fed back from the vision system. Firstly, the real-time image captured by the calibrated industrial camera is filtered to improve the image contrast and then to extract the landmark outline. Secondly, the contour features are extracted using two different linear detection techniques, and the real-time offset angle and offset amount of the spreader are calculated and fed back to the control system. Finally, by building an experimental platform to verify the stability, real-time and accuracy of the vision system, it is concluded that the detection method can fully meet the actual port transportation needs and can replace manual detection.