Varieties discrimination of accumulation teas based on BP neural network model
ZHANG Zhi-feng
LI Shi-hai
TANG Yi-ming
QIAO Lin
WU Fan
ZHAI Yu-sheng
Abstract：A rapid and nondestructive method to discriminate the varieties of accumulation teas was proposed.The color images of the detecting teas with natural grain accumulation captured by CCD were preprocessed with filter, and analyzed by gray level co-occurrence matrix to gain texture feature of tea.Three characteristic parameters from principal component analysis were the input parameters of BP neural network model to set up pattern recognition model.The experimental results showed that the predictive accuracy was 93.8%for unknown 32 forecast samples. the system and method can meet the requirement of tea production and trade circulation.The method provided a kind of recognition technology to realize the tea varieties rapid nondestructive identification and improve the recog-nition accuracy in the process of production,processing and trade of the tea.
Keywords：gray level co-occurre-nce matrixBP neural network modelprincipal component analysistea nondestructive identifi-cation
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 103-108 )