基于GWO优化SVM的小麦籽粒优劣分级研究

Study on the Classification of Wheat Grain Quality Based on GWO Optimized SVM

小麦籽粒优劣不仅是产量及品质的重要决定因素,也是育种适应性的综合指标。为了提高小麦籽粒优劣分级的准确率,同时克服神经网络中存在的收敛速度慢、容易陷入局部极值等缺陷,提出一种灰狼算法(GWO)优化支持向量机(SVM)的小麦籽粒优劣分级方法,以航麦8805为研究对象,利用图像处理技术对小麦籽粒图像进行预处理并提取小麦籽粒的形态、颜色和纹理等21个特征。然后采用灰狼算法对支持向量机的两个参数(c、σ)进行优化,建立GWO-SVM模型,从而对小麦籽粒进行优劣分级。与其他算法相比,GWO优化SVM的算法对小麦籽粒的分级准确率有明显的提高,对小麦籽粒优劣分级的准确率可达到95.08%。

Wheat grain quality is not only an important determinant of yield and quality but also a comprehensive indicator of breeding adaptability.In order to improve the accuracy of wheat grain grading,at the same time to solve the problem of local optimal solution and slow convergence rate existed in artificial neural networks,we presented a method based on support vector machine(SVM)with gray wolf algorithm optimization(GWO)for grading of wheat grain.First of all to wheat 8805 as the research object,using the image processing technology of wheat grain image preprocessing and extract the grain morphology,color,texture,a total of 21 characteristics.Then,GWO was used to optimize the two parameters(c,σ)of the SVM,and the GWO-SVM model was established to grading of wheat grains.Comparing the GWO optimized SVM algorithm with other algorithms,the results showed that the accuracy of the GWO optimized SVM algorithm for wheat grain classification can reach 95.08%,which is significantly higher than other algorithms.