Image recognition technology and its application in the interactive design of commercial self-service robot

Robot technology is a high-tech integrated with many disciplines, and it is very active in contemporary research. It is also an important symbol of a country's industrial automation level. With the development of science and technology, the connotation of robots is constantly enriched. The introduction and popularization of commercial service robots is both a trend and an inevitable trend. The service robot industry is developing rapidly, and the competition among industries is also fierce. If the service robot enterprise wants to have certain competitiveness in the market, it must have the technology commercialization ability and transform the enterprise technology into commercial value. Through the research of the existing commercial service robot products, this paper summarizes the product characteristics of the commercial service robot. Then, aiming at the application of image recognition technology in the interactive design of commercial self-service robots, a tracking method based on deep learning is proposed. Using the pre-trained deep convolution network as the feature extraction unit, and converting the tracking into template matching, has the advantage of balancing the tracking accuracy and speed. In this paper, a new template updating strategy is proposed, which can judge the timing of target template feature updating, so as to achieve effective template updating and improve the robustness of the tracker. The experimental results show that the model has better performance.