【摘要】 针对工业产品识别准确度的不足，提出一种基于5G网络的工业产品识别系统框架。在传统卷积神经网络CNN基础上，提出DNN深度神经网络算法，并引入控制参数k_p实现采集数据的变换特征映射;基于5G网络传输和加密，实现实时数据传输和保密通信;基于云平台，使用Pytorch进行物体识别的匹配运算;最后开发了一套基于5G网络的工业产品识别与溯源平台，验证本设计的实用性。结果表明，本系统对于工业产品识别具有很高的准确率和识别效果。更多还原

【Abstract】 In view of the Industrial Product Identification Accuracy,this paper presents a framework of industrial product identification system based on 5G network.Based on the traditional convolutional neural network (CNN),DNN (deep neural network) algorithm is proposed,and the control parameter k_p is introduced to realize the transformation feature mapping of the collected data.Finally,a set of industrial product identification and traceability platform based on 5G network is developed to verify the practicability of the design.The results show that the system has a high accuracy and recognition effect for industrial products.更多还原