基于Object Detection API的物流单元货架目标检测  

Object Detection of Logistics Unit Shelf Based on Object Detection API

作　　者：龙健宁 刘斌[1] 龚德文 LONG Jian-ning;LIU Bin;GONG De-wen(National Engineering Research Center of Novel Equipment for Polymer Processing,Key Laboratory of Polymer Processing Engineering of Ministry of Education,Guangdong Provincial Key Laboratory of Technique and Equipment for Macromolecular Advance Manufacturing,South China University of Technology,Guangzhou 510641,China;Guangdong Changheng Intelligent Technology Co.,Ltd.,Dongguan 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]广东昌恒智能科技有限公司,东莞523841

出　　处：《自动化与仪表》2020年第9期46-50,55,共6页Automation & Instrumentation

摘　　要：随着人工智能的兴起,深度学习的方法已经被广泛地应用到各类图像目标的检测当中,并在复杂环境下取得了良好的效果。针对物流仓储环境,该文基于开源框架Tensorflow上的库Object Detection API,选择了Faster R-CNN算法和SSD-MobileNet算法,分别对物流单元货架上摆放的物流周转箱进行目标检测。实验结果表明,相比于Faster R-CNN算法,SSD-MobileNet算法能够同时满足实时性与准确率的要求。将训练所得的SSD-MobileNet模型移植到QT平台,设计了物流单元货架目标检测界面。With the rise of artificial intelligence,deep learning method has been widely used in all kinds of image object detection,especially good results has been achieved in the complex environment.Based on the library Object Detection API in Tensorflow,Faster R-CNN algorithm and SSD-MobileNet algorithm are selected in this paper to detect the object of the logistics turnover box on the logistics unit shelf.The experimental results show that compared with Faster R-CNN algorithm,SSD-MobileNet algorithm can meet the requirements of real-time and accuracy at the same time.The SSD-MobileNet model is transplanted to QT platform,and the object detection interface of logistics unit shelf is designed.

关 键 词：深度学习 物流单元货架 目标检测 Faster R-CNN算法 SSD-MobileNet算法 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]