Combining genetic algorithm and SVM for corn variety identification

In this article, a real-time, accurate and objective identification of different varieties of corn seeds is proposed, which is a large number of original features, contained color, texture and shape features, were extracted from corn seed images. Then, genetic algorithm and support vector machine (SVM) were used to select important ones and determine species. The proposed methods have optimized varieties recognition algorithm based on machine vision, which also improved the accuracy and achieved the best performance percentage of 94.4%. Basically, the average consumption time for every seed is 0.141s.