Research on Robot Cooperative Intelligent Manufacturing System Based on Machine Vision

This research focuses on the tracking of complex dynamic targets by machine vision in intelligent manufacturing, and combines DCF and Transformer technologies to innovate the DTF model. By integrating localization coding, mask conversion and local localization functions, the model can effectively deal with multi-scale tracking problems and show high stable tracking performance in simulated intelligent manufacturing environments. The DTF achieved a Precision of 83.73% and an AUC of 76.14% in the benchmark tests, surpassing comparable advanced systems. The ablation experiments highlight the indispensability of each module for performance. This research provides an advanced visual tracking tool for robot collaboration and promotes the intelligent process of intelligent manufacturing.