Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

This article describes a study of the feasibility of identifying tea varieties using computer vision. Five varieties of Chinese green tea were used in the experiment: Biluochun, Tunchaoqing, Maofeng, Queshe, and Maoshanchangqing. Tea images were grabbed by a computer vision system by spreading tea leaves uniformly. Twelve color feature variables were extracted from the RGB and HSI color spaces, and 12 texture feature variables were extracted based on statistical moment measurement and spectral measurement. Linear discriminant analysis (LDA) was applied to build the identification model based on principal component analysis (PCA), and the number of principal components was optimized in building the model. The performance of the LDA model was optimal when the number of principal components was 11; the identification rate was 100% for the training set and 98.33% for the prediction set. The experimental results showed that five varieties of tea could be successfully identified by extracting suitable features.