Similarity Analysis Of The Three-Dimensional Point Cloud Based On Eigenvector Of The Subspace Of Dairy Cow’s Breast

Because milk is an important food source of national, so the research with the dairy cow-related has been subject to a high degree of attention. The body structures of the high-yielding dairy cows and low-yielding dairy cows, especially in the body structures with breast related have significant differences. after a long period of research and practice, the importance of body structures and milk production performance has been referred to the same status. However, due to cow breast shape evaluation is done manually, so time-consuming and accuracy is not high, Meanwhile because manual inspection often need contacts with the cows, so cows can cause stress to reflect, and affect their production. Nowadays, Analysis of the features Cow’s breast shape based on computer vision is expected to become an effective means to solve these problems.Therefore, this research using computer vision and reverse engineering, the cow’s breast shape characteristics of measurement and analysis, to explore efficient, non-contact method of assessing cow breast shape characteristics. Compared with manual measurement, the method is simple and fast, efficient, avoid subjective factors, and because no physical contact with cows, so cows can minimize the stress response. similarity analysis algorithm of the three-dimensional point cloud proposed in this paper, and the proposed method is expected to summarized the characteristics of high producing dairy cows breast shape, and shape characteristics of milk performance analysis to provide an effective means.This paper presents a method of similarity analysis algorithm of the three-dimensional point cloud, which is based on eigenvector of the subspace. First of all, we obtain the three-dimensional point cloud data of two objects and standardize the position of them. And then, the two three-dimensional point clouds are divided into several subspace by using the minimal spatial segmentation algorithm. Thirdly, we calculate the eigenvector of subspace,which should be divided into two steps:the first step is to calculate distance and angle from the centroid to the subspace surface, the next step is to compute the new eigenvector on the basis of vector space, which is composed of the distance and angle in step one. This research method takes the advantage of small data in quantity and high precision in calculation because the eigenvector of subspace, which can describe the three-dimensional characteristics, is taken as the basis of similarity measure. The experiment shows that the algorithm can quantitatively analyze the similarity of two three-dimensional objects.Currently, the laboratory has completed the phase of the study, the next stage of plans to collect a large number of dairy cows in vivo morphological characteristics, internal organizational structure characteristics, and milk performance in-depth study.