Body weight estimation of yak based on cloud edge computing

Abstract In stock farming, the body size parameters and weight of yaks can reasonably reflect growth development characteristics, production performance genetic characteristics yaks. However, it is difficult for herders to measure by traditional manual methods. Fortunately, with edge computing, use mobile devices estimate yak’s weight. The purpose this paper provide a machine vision-based yak estimation method equipment establish comprehensive display system based on user’s in order maximize convenience herdsmen’s work. our method, set image foreground extraction measurement point recognition algorithm suitable were developed obtain image, ratio between sizes was transmitted cloud server. Then, estimated using data mining constantly displayed system. Twenty-five different age groups randomly selected from herd perform experiments. experimental results show that segmentation good boundary, has accuracy stability. average error values actual measured height, oblique length, chest depth, cross height 1.95%, 3.11%, 4.91%, 3.35% 7.79%, respectively. Compared end improve efficiency, facilitate herdsmen breed yaks, reduce stimulation lay solid foundation fine breeding Sanjiangyuan region.