山西农业大学学报（自然科学版）2019，Vol.39Issue(2)：86-92,7.DOI:10.13842/j.cnki.issn1671-8151.201810033
基于机器视觉的武夷岩茶做青程度识别研究
Study on the color degree recognition of Wuyi Rock Tea with machinery vision
摘要
Abstract
[Obj ective]The obj ective of this study was to analyze the correlation between the color change of tea leaves after sun withered and the degree of green processing using machinery vision technology.[Methods]Color images were captured by an image acquisition system in green processing phase,and the color change of tea leaves was monitored by RGB and HSI models combined with self-organized artificial neural network technology to classify different degrees of leave green processing.[Results]The classification accuracy rates of training set and testing set of 140 samples ex-pressed by RGB component and HSI component were 81.64%,78.6% and 76.5%,73.8%,respectively.The proba-bility (P)of significance testwas less than 0.05,and the coefficient of determination R2 was equal or greater than 0.879. The color component was highly correlated to the degree of green processing.[Conclusion]The self-organized artificial neural competition network model can be used to predict the degree of Wuyi Rock Tea green processing,and to provide a reference for monitoring the process of Oolong tea green processing by machinery vision in the future.关键词
武夷岩茶/机器视觉/神经网络/做青程度/识别Key words
Wuyi Rock Tea/Machine vision/Neural network/Degree of greening/Recognition分类
信息技术与安全科学引用本文复制引用
吴薇,王宜怀,程曦,郑慕蓉..基于机器视觉的武夷岩茶做青程度识别研究[J].山西农业大学学报（自然科学版）,2019,39(2):86-92,7.基金项目
贵州省科技成果应用及产业化计划项目(黔科合成果[2018]4311) （黔科合成果[2018]4311）
贵州省亚热带作物科技创新人才基地建设经费[黔人领发(2016)22号] （2016）
