Identification and Grading of Tea Using Computer Vision

Abstract

In order to objectively evaluate the tea's quality, this article uses the computer vision to classify and grade the tea based on the parameters of the tea's color and shape. Images of tea leaves were directly acquired by digital devices (e.g. digital camera), and then the color feature parameters were extracted on the basis of the HSI model and shape feature parameters were extracted on the basis of binary images after being converted and pre-processed. The automatic identification of tea was achieved by using the genetic neural-network. According to the experimental results, this method could achieve a better effect of identification with eight parameters of color and shape, and the results of computer-based tests were highly consistent with those of manual tests.