摘要
随着信息技术的不断发展,零售行业大量使用扫码方式进行商品的识别,十分快捷方便。但是对于糕点、面包等无法附加条形码的商品,只能人工进行商品的识别。使用商品自动识别技术可以提高商品结算的效率,降低人力成本。因此,本文通过在小型本地设备上部署轻量化的深度学习多目标识别算法,可以快速地进行烘培类商品的自动识别。经过多家糕点店的实际测试验证,本系统能够准确、快速地对放置于购物盘中的商品进行识别。
With the continuous development of information technology,code scanning is widely used in the retail industry for commodity identification,which is very fast and convenient.However,for cakes,bread and other commodities that cannot be attached with bar codes,they can only be identified manually.The use of automatic commodity identification technology can improve the efficiency of commodity settlement and reduce labor costs.Therefore,by deploying lightweight deep learning multi-target recognition algorithm on small local equipment,this paper can quickly carry out the automatic recognition of baking goods.After the actual test of many pastry stores,the system can accurately and quickly identify the goods placed in the shopping tray.
作者
李良熹
荣进国
LI Liangxi;RONG Jinguo(Chongqing Technology and Business Institute,Chongqing 401520,China)
出处
《信息与电脑》
2021年第13期156-158,共3页
Information & Computer
基金
重庆工商职业学院校级科研项目“基于边缘计算的智能机器人实时多目标跟踪系统关键技术研究”(项目编号:NDYB2020-15)
重庆工商职业学院人才引进科研项目“基于深度学习的实时目标检测技术研究”(项目编号:ZZ2020-05)。