Method for determining sexual dimorphism of embryos of batch of poultry eggs before incubation, involves using selected data in machine learning process applicable in solving problem of binary classification for small samples with large dimension of initial set of features
2023-03-09
专利权人AS RUSSIA SIBERIAN FEDERAL RES CENT AGRO (ASSI-Non-standard)
申请日期2023-03-09
专利号RU2023105569-A
成果简介NOVELTY - The method involves calculating the perimeters and sums, differences and ratios to each other on sections outside the inscribed circle in pixels the areas, at the acute and blunt ends of the image of the egg. The values are divided into a system of m radius vectors and the modules are determined in pixels from the selected center of the figure to the boundary of the contour of the images of specific egg. Four new images are formed from each sector, to determine the halves of the transverse and longitudinal dimensions of the image of a specific egg. The areas and perimeters are calculated, then the data obtained above for determining the parameters of egg shapes are subjected to static analysis and the data are selected for which the level of significance in selected size group is the lowest. The selected data are used in the machine learning process applicable in solving the problem of binary classification for small samples with a large dimension of the initial set of features. USE - Method for determining sexual dimorphism of embryos of a batch of poultry eggs before their incubation.
IPC 分类号A01K-043/04 ; A01K-045/00
国家俄罗斯
专业领域农业科学
语种英语
成果类型专利
文献类型科技成果
条目标识符http://119.78.100.226:8889/handle/3KE4DYBR/22197
专题中国科学院新疆生态与地理研究所
作者单位
AS RUSSIA SIBERIAN FEDERAL RES CENT AGRO (ASSI-Non-standard)
推荐引用方式
GB/T 7714
ALEYNIKOV A F,OSIPENKO I V,YAKOVINA I N,et al. Method for determining sexual dimorphism of embryos of batch of poultry eggs before incubation, involves using selected data in machine learning process applicable in solving problem of binary classification for small samples with large dimension of initial set of features. RU2023105569-A[P]. 2023.
条目包含的文件
条目无相关文件。
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。