Abstract: Aimed at rape nutrient deficiency，we applied support vector machine in rape nutrient deficiency diagnosis. Firstly，we determined the feature value for support vector machine classification，chose the RGB and HSV color space as color features，and chose the mean and variance of energy，entropy，contrast，correlation as texture features. Then，the support vector machine was applied to classify pattern recognition and was compared to BP neural network. The results showed that the support vector machine was superior to BP in classification performance. Finally， through genetic algorithm to optimize the parameters of support vector machine，the final classification accuracy was improved.