基于粒子群算法和神经网络的大豆识别研究

Soybean Classification Based on Particle Swarm Optimization and Neural Network

在分析大豆表面颜色特征的基础上,提出了利用神经网络和大豆表面颜色特征对大豆进行分级的方法。选取了大豆图像的6种颜色特征值作为神经网络的训练样本,并尝试利用粒子群优化算法与BP结合算法训练网络。仿真结果表明,提出的方法取得很好的效果。

Based on analysis of sofbean surface color features, a classification method by utilizing neural network and soybean surface color features is presented as follows： Six color feature data are chosen as the input of a classifier, and a new algorithm which combines particle swarm optimization with BP is used to train neural network. Simulation results prove the effectiveness of the approach introduced in this paper.