【摘要】 为解决大型通信设备的部件识别问题,提出了一种将SIFT特征和SVM相结合的分类方法。首先,通过SIFT算法得到样本图片的feature集即特征向量,并通过K-means聚类算法得到中心矩阵,再分别将所有feature集与中心矩阵作欧氏距离计算并统计最小值位置,即可得到输入数据;然后,采用以高斯径向基函数为核函数的一对多SVM分类器进行训练;最后,对新的输入数据进行检测,并得到输出结果。试验结果表明,基于SVM的分类方法能够有效提高分类正确率,并达到92%以上。更多还原

【Abstract】 In order to solve the recognition problem of large-scale communication equipment components,the classification method combining SIFT and SVM is proposed in this paper. Firstly,it gets the feature collection of sample images by SIFT algorithm,and gets the center matrix by K-means clustering algorithm,and then Euclidean distance calculation is made between the feature collection and the center matrix,and it counts the position corresponding the minimum distance,so the input data is gained. Then,the SVM classifier is used for training whose kernel function is Gaussian radial basis function. Finally,it detects the new input data and gets the output. The test results show that the SVM method can effectively improve the classification accuracy rate,and it is more than 92%.更多还原