Open AccessRecognition of milk somatic cells based on dichotomy model
Author(s) -
Lei Chen,
Xiaojing Gao,
Heru Xue
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/569/2/022034
Subject(s) - somatic cell , artificial intelligence , pattern recognition (psychology) , artificial neural network , naive bayes classifier , feature (linguistics) , texture (cosmology) , computer science , mathematics , image (mathematics) , biology , support vector machine , genetics , linguistics , philosophy , gene
The number of various milk somatic cells can be calculated accurately by using image processing technology. In order to identify four kinds of cells,a method of milk somatic cell recognition based on dichotomy model was proposed. Firstly, eight typical morphological features were extracted from cells. Some morphological features were selected by feature optimization and Bayes classifier was used twice to carry out two binary classification. The texture feature described by HOG and BP neural network were uesd on the last classify. Finally, 96.02% of the average classification accuracy was achieved in the actual experiment of milk somatic cell image data.