Abstract: In order to carrying on a digital, normalize, quantitative and undestroied method for plant disease discriminating, two kinds of familiar cucumber diseases (cucumber anthracnose and cucumber brown speck) were researched in the method of picture processing. Color characteristic parameters coming from different periods of disease simple were used to classification and identification . Using stepwise discriminant analysis to select significant parameters, and the bavesian classifier method was proposed to classify cucumber anthracnose, cucumber brown spot and healthy region. The result shows that the correct discrimination of cucumber anthracnose, cucumber brown spot and healthy region were 96.67%, 93.33% and 100%, respectively. For the test set，the correct discrimination of cucumber anthracnose, cucumber brown spot and healthy region were 83.33%, 80.00% and 100%, respectively. The results show that it is feasible to identify and classify plant diseases using color image statistical characteristics.

Key words: Cucumber, Plant disease, Color image, Color, Statistical characteristics, Quantitative discrimination