Animal shape phenotype automatic measuring method

Abstract

Disclosed is an animal shape phenotype automatic measuring method. An animal shape phenotype is automatically measured through template deformation. The method comprises the following steps: establishing a database, i.e., establishing a template model database storing three-dimensional stereo templates of different animals in different growth periods, and a shape phenotype measuring definition database storing shape phenotype index measuring definitions measuring different animal shape phenotype indexes; coupling, i.e., according to the three-dimensional stereo templates in the template model database, comparing input animal data models, obtaining three-dimensional stereo templates which are closest to the types of the currently input animal data models and the growth periods as the coupling templates of the currently input animal data templates; deformation, i.e., according to the input animal data models, deforming the coupling templates until the coupling templates are close to the currently input animal data models in terms of body size; and calculating, i.e., according to the shape phenotype measuring definitions in the shape phenotype measuring definition database and the deformed coupling templates, obtaining the shape phenotype indexes of the input animal data models. The animal shape phenotype automatic measuring method improves the automation degree of animal shape phenotype measurement.

Description

technical field [0001] The invention relates to the field of three-dimensional (3D) computer vision, in particular to an automatic measurement method for animal shape phenotypes based on template deformation. Background technique [0002] In the agricultural production process, the application of animal shape phenotypes such as horses, cattle, sheep and pigs includes breeding, weight estimation and body measurement. These applications all involve various indicators for measuring animal shape phenotypes, but the large number of various evaluation indicators required leads to a large amount of data processing workload. Therefore, the automatic high-throughput acquisition of animal 3D shape phenotype-related indicators has certain production significance. [0003] In the actual production at present, most of them are measured manually by measuring tools such as half-circle gauges, calipers, and tapes, or directly observed and scored by experts. The acquisition of evaluation i...

AI Technical Summary

Problems solved by technology

The requirements for animal pose or collection environment are relatively strict, especially in the actual domestic production environment, which is difficult to meet

Therefore, in the current practical application, the manual observation and evaluation of experts is still the main method, and the two-dimensional image technology itself cannot obtain the indicators related to the shape of the animal body surface

Images