论文部分内容阅读
针对油菜常见的缺素现象,提出将支持向量机应用于油菜缺素种类识别。首先,确定支持向量机分类过程中所用的特征值,选择RGB和HSV颜色空间中的分量作为颜色特征,选择能量、熵、对比度、相关性的均值和方差作为纹理特征;其次,将支持向量机用于分类模式识别,并与神经网络分类识别进行比较,仿真结果表明:支持向量机的分类精度高,性能更好;最后,通过遗传算法对支持向量机参数进行优化,可以看到最终的分类准确率有所提升,起到了优化的效果。
Aimed at the common deficiency of rapeseed, a new support vector machine (SVM) was proposed to identify the species of rapeseed. Firstly, the eigenvalues used in SVM classification are determined, the components in RGB and HSV color space are selected as the color features, and the energy, entropy, contrast, mean and variance of correlation are selected as texture features. Secondly, the support vector machine Which is used for classification pattern recognition and compared with neural network classification and recognition. The simulation results show that SVM has high classification accuracy and better performance. Finally, the genetic algorithm is used to optimize SVM parameters to find the final classification Accuracy has improved, played an optimization effect.