The invention discloses an intelligent detection method and system for a rice variety. The method comprises the following steps that: carrying out classified labeling on the varieties of rice samples;for a certain variety, taking the variety as a positive sample of neural network training, and taking the samples of reset varieties as the negative samples of the neural network training to train the neural network; when all verities finish being trained, obtaining a classifier; and inputting a sample to be detected into the classifier, and utilizing the classifier to detect the rice variety. Byuse of the method, all appearance characteristics of the detected rice can be described and input into the neural network, the neural network is trained, and the trained neural network can accuratelydetect the rice variety and mixing according to sensitive characteristics, including the "grain shape" and the "appearance", of the rice.