应用卷积网络及深度学习理论的羊绒与羊毛鉴别

为解决羊绒与羊毛纤维图像难以鉴别的问题,提出一种基于卷积网络和深度学习理论的鉴别方法。使用sigmoid分类器将卷积深度网络提取的纤维图像特征进行粗分类,根据验证集合验证结果并记录网络的最优权重。根据整体的分类网络所获取的权值,对每张样本图片使用改进的局部增强整体的网络模型提取局部特征,并对局部特征和整体特征进行融合,根据这些融合特征建立新的分类网络。在此基础上,使用鄂尔多斯标准羊绒与羊毛数据集对网络进行50轮次的迭代训练,得到的最优准确率达92.1%。实验结果表明:采用深度卷积网络表征纤维,并对羊绒羊与毛纤维图像进行分类的方法,能够有效解决羊绒、羊毛等类似纤维鉴别问题;若用于商业检测,还需更多数据集的验证。

Identification of cashmere and wool based on convolutional neuron networks and deep learning theory

In order to solve the problem of the difficulty in identification of cashmere and wool fiber images,a novel identification method based on convolutional neuron network (CNN) and the deep learning theory was proposed.A sigmoid classier was used to carry out coarse classification on fiber image features extracted by the CNN.The results were verified according to the validation datasets,and the optimal weights of the network was recorded.Part-level features of each sample image are extracted by an improved part-level augment object-level network based on previously obtained parameters.In addition,the part-level and object-level features were fused,and a new network model was established based on the fused features.On this basis,Ordos's standard cashmere and wool dataset was used to train the network for 50 times,and the best accuracy is 92.1％.The experiment results demonstrate that the method for classifying cashmere and wool images,based on the CNN features,can be applied to cashmere and wool or discriminate similar fibers identification successfully.However,a large number of sample sets and validations are required for commerical use.