A Volume Assessment Model of Bulk Cargo Terminal Yard Based on Semi-global Stereo Matching

When the volume is evaluated by the method that relies only on monocular vision, only two-dimensional plane information can be obtained, and the reliability of the evaluation result is low. Therefore, a bulk cargo terminal yard volume evaluation model based on semi-global stereo matching is proposed. We use industrial cameras to collect images, which are processed by stereo correction, homogeneous coordinate transformation and pixel mapping in turn. The semi-global stereo matching algorithm is used to determine the disparity value through path cost aggregation and WTA algorithm. The three-dimensional point cloud data is obtained through the relationship between disparity and depth and coordinate transformation, and then the three-dimensional model is constructed through denoising, hole filling and surface reconstruction. Finally, the tetrahedral unit is divided to evaluate the bulk cargo terminal yard volume. The experimental results show that this method significantly improves the evaluation accuracy, reduces the mean square error, and enhances the reliability of the evaluation results.