基于卷积神经网络的物流货物图像分类研究  

Research on Goods Image Classification Based on Convolutional Neural Network in Logistic Environment

作　　者：刘斌[1] 龙健宁 程方毅 龚德文 Liu Bin;Long Jianning;Cheng Fangyi;Gong Dewen(Key Laboratory of Polymer Processing Engineering of Ministry of Education//Guangdong Provincial Key Laboratory of Technique and Equipment for Macromolecular Advance Manufacturing//National Engineering Research Center of Novel Equipment for Polymer Processing,Guangzhou 510641,China;Guangdong Changheng Intelligent Technology Co.,Ltd.,Dongguan,Guangdong 523841,China)

机构地区：[1]华南理工大学聚合物成型加工工程教育部重点实验室//广东省高分子先进制造技术及装备重点实验室//聚合物新型成型装备国家工程研究中心,广州510641 [2]广东昌恒智能装备科技有限公司,广东东莞523841

出　　处：《机电工程技术》2021年第12期79-82,175,共5页Mechanical & Electrical Engineering Technology

摘　　要：基于卷积神经网络(CNN)针对物流环境下货物的图像分类问题进行了研究。首先,在实际物流环境下收集了13种货物的ROI图像,并通过每隔10°旋转的方式来扩充数据集以防止过拟合现象的发生;然后,在考虑了实际硬件条件的情况下构建了轻量级CNN,并进行了基于自建数据集的训练,训练实验发现,轻量级CNN模型具有很快的收敛速度并在验证集取得了100%的准确率;最后,研究了旋转对货物图像分类性能的影响,并进行了可视化分析,验证了CNN对旋转操作基本不具备一致性。The classification of goods images in logistics environment based on convolutional neural networks has been studied.First,ROI images of 13 kinds of goods in logistics environment were collected and the data set was expanded by rotating every 10° to prevent overfitting.Then,a lightweight CNN was constructed and trained based on self-built data set considering the actual hardware conditions,and the training experiment showed that the model converged quickly and achieved 100% accuracy in the validation set.Finally,the effect of rotation on the classification performance of goods images was studied and performed a visual analysis,the result showed that CNN had poor invariance to rotation operation.

关 键 词：卷积神经网络 图像分类 自动识别 旋转不变性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]