Performance analysis of robotic arm visual servo system based on BFS-canny image edge detection algorithm

Visual servoing technology is an important direction for realizing the intelligence of robotic arms, which enhances the flexibility and robustness of robotic arms by introducing visual feedback information. Image processing is the core part of this technology, but its real-time performance and stability are currently facing significant challenges. To solve the above problems, this study designs an image edge detection algorithm by combining breadth first search algorithm, Canny algorithm, Harris algorithm, and parallel processing strategy, and establishes a robotic arm visual servo system. The results showed that with the continuous increase of data volume, the performance of the proposed algorithm was still superior to other image edge detection algorithms throughout the entire process, and the accuracy, recall, and F1 score indicators exceeded 95%, 86%, and 90%. Under the image size of 4096*2160, the research algorithm could achieve a computational efficiency of 110FPS, and the average running time in the test dataset was at least 30.28ms. In practical application, the research method could converge the tracking error to a smaller range and greatly improve the dynamic tracking performance. The above data indicate that the research method can effectively improve the performance and reliability of the robotic arm visual servo system, and achieve perception of complex application environments.