Construction and Simulation of Deep Learning Algorithm for Robot Vision Tracking

Abstract

As one of the indispensable basic branches of computer vision, visual object tracking has very important research value. Therefore, a deep learning based on robot vision tracking is evaluated. Based on the basic principles of target tracking and search principle, a deep learning algorithm for visual tracking is constructed, and finally, evaluated, and simulated. The results showed that the accuracy rate increased from 90.9% to 90.13% after the addition of channel attention mechanism module. Variance was reduced from 3.78% to 1.27%, with better stability. The EAO, accuracy, and robustness of the algorithm are better than those without significant region weighting strategy. The strategy of using the improved residual network SE-ResNet network to extract multiresolution features from the correlation filtering framework is effective and helpful to improve the tracking performance.

An improved braitenberg policy for object tracking

“…In the field of traffic monitoring, the color characteristics of vehicles are detected and tracked to achieve traffic flow statistics, traffic violation detection, and other functions [2]. In the field of autonomous driving, deep learning tracking is widely used to achieve tracking and recognition of targets such as vehicles, pedestrians, and traffic signs by analyzing and predicting sensor data in real time using deep learning models [3]. Various methods have been used to achieve tracking such as color, feature point, motion model or deep learning based tracking methods.…”

“…The suggested approach produces good results and is resistant to changes in the shape, texture, illumination, and occlusion of objects. However, a significant amount of training data and computational power are needed [3].…”