基于离散粒子群优化算法的储粮害虫特征选择

Feature Selection for the Stored-grain Insects Based on Distributed Particle Swarm Optimization

储粮害虫特征选择是粮虫自动检测系统的一个关键环节。为此,提出了把离散粒子群优化算法应用到粮虫的特征选择。该算法从粮虫的17维形态学特征中自动选择出面积、周长等7个特征的最优特征子空间,采用参数优化后的SVM分类器对90个粮虫样本进行分类,识别率达到95%以上,证实了基于离散粒子群优化算法的粮虫特征选择是可行的。

The feature selection of the stored - grain insects is a important part for the automatic detection system. The distributed particle swarm optimization algorithm wais applied to the feature selection. The algorithm selected seven features that composed the optimal feature space from the 17 morphological features, such as area and perimeter. The ninety image samples of the stored- grain insects in grain- depot were automatically recognized by the support vector machine classifier using the optimized parameters, and the correct identification ratio was over 95 %. The experiment showed that it was practical and feasible.