An intelligent vision system for the detection and classification of heterogeneous objects moving on a belt conveyor

In this paper, a newly developed author’s vision system allowing identification of packages moving on the conveyor belt and their classification is presented. An important aspect is that the mentioned objects move at relatively high speeds, which is challenging for any vision system. The methodology used in the plant is based on a hybrid combination of algorithmic determination of the features of a given object with artificial neural networks performing the classification process. An essential aspect of the work is using the advanced camera, which provides three-dimensional image analysis. The obtained verification results confirm the system’s usefulness not only for research purposes but even for implementing industrial automation tasks.