基于卷积神经网络的SVM分类算法在图像分类中的应用

Application of SVM Classification Algorithm Based on Convolutional Neural Network in Image Classification

利用卷积神经网络强大的自学能力,训练合适的CNN来提取图像特征信息,利用RBF函数作为支持向量机的核函数,并结合粒子群算法优化SVM参数,完成图像分类的混合算法。针对乳腺组织的病理图像分类性能的实验分析,给出了混合分类算法的优越性。

Using the powerful self-learning ability of convolutional neural network,this paper trains a suitable CNN to extract feature information,selects RBF function as the kernel function of support vector machine,and completes the mixed type classification algorithm by optimizing the parameters of particle swarm optimization algorithm.Through the experimental analysis of the classification performance of pathological images of breast tissue,the advantages of the new classification algorithm are obtained.