Abstract: The vision of live pigs and its auxiliary information is a very important basic information for the identification of the health status of live pigs. It can not only reflect the growth of live pigs in real time, but also is an important basis for the diagnosis and treatment of live pig diseases. With the continuous development of image processing and machine learning technology, computer vision technology with no damages, high efficiency, low cost, automation and real-time has been gradually popularized in large-scale pig farms. The author summarized the typical research work on computer vision technology in pig healthy breeding in recent 20 years at home and abroad. On the basis of describing the basic concept and composition of computer vision system, the application and algorithm of vision technology in pig target extraction, identity recognition, motion monitor, behavior recognition, quality estimation and diseased area detection were reviewed. The application status of pig disease diagnosis, growth prediction and precise nutrition based on computer vision measurement were analyzed and discussed, in order to evaluate the overall application of computer vision technology in pig breeding and the existing problems, the purpose is to provide reference for the research and development of computer vision technology in pig healthy breeding.