文章基本信息

标题：Label number Recognition Based on Convolutional Neural Networks in industrial products
本地全文：下载
作者：Yao Yang ; Na Qin ; Deqing Huang 等
期刊名称：IFAC PapersOnLine
印刷版ISSN：2405-8963
出版年度：2019
卷号：52
期号：24
页码：207-212
DOI：10.1016/j.ifacol.2019.12.409
语种：English
出版社：Elsevier
摘要：Aiming at the identification of a certain kind of industrial black material product, this paper proposes a method based on convolutional neural network (CNN) for digital identification of product labels. The platform of image acquisition is set up first, then the digital region is segmented through image processing algorithm and data set is built on it. Finally, the visual geometry group (VGG16) model of convolutional neural network is used to realize the identification of digital labels. Compared with the nearest neighbor based on local binary patterns histograms (LBPH-NN) algorithm and the support vector machine (SVM) algorithm, the performance of CNN is better comprehensively. This research has a good practical significance in the field of industrial production.
关键词：KeywordsImage preprocessingDigital extractionConvolutional neural networkVGG16Fine-tuning