卷积神经网络物体检测算法在物流仓库中的应用

doi:10.19678/j.issn.1000-3428.0047321

计算机工程 ›› 2018, Vol. 44 ›› Issue (6): 176-181. doi: 10.19678/j.issn.1000-3428.0047321

卷积神经网络物体检测算法在物流仓库中的应用

李天剑 ¹,黄斌 ²,刘江玉 ¹,金秋¹

1.北京信息科技大学机电工程学院,北京 100192; 2.北京航空航天大学自动化科学与电气工程学院,北京 100191

收稿日期:2017-05-23 出版日期:2018-06-15 发布日期:2018-06-15
作者简介:李天剑(1969—),男,副教授、博士,主研方向为图形图像处理、机器人技术、计算机视觉;黄斌,博士研究生;刘江玉、金秋,硕士研究生。
基金资助:
北京市科技计划项目(Z171100000817006)。

Application of Convolution Neural Network Object Detection Algorithm in Logistics Warehouse

LI Tianjian¹ , HUANG Bin ² ,LIU Jiangyu ¹,JIN Qiu¹

1.School of Mechanical and Electrical Engineering,Beijing Information Science and Technology University,Beijing 100192,China; 2.School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China

Received:2017-05-23 Online:2018-06-15 Published:2018-06-15

摘要/Abstract

摘要： 针对传统物体检测算法在复杂环境下检测准确率较低的问题,提出一种新的托盘检测算法。采集真实仓库中包括人和托盘的大量图片进行标注,构建物流仓库的托盘数据库,并将单次多箱探测器检测算法中的基础网络改进为DenseNet网络,利用所标注的托盘数据库进行训练和测试。在测试阶段,结合不同分辨率的多尺度特征图,以增强网络对被检测物体的适应能力,并使用单一网络实现检测任务。实验结果表明,与YOLO算法相比,该算法检测准确率提高了6.1%。

关键词: 物体检测, 托盘检测, 卷积神经网路, 深度学习, 稠密连接卷积神经网络

Abstract: In complex environments,the accuracies of traditional object detection algorithms is too low.A new pallet detection algorithm is proposed.A large number of pictures containing people and pallets in real warehouses are collected and annotated to building the pallets dataset for training.In the paper,the based network of the improved algorithm based on Single Shot MultiBox Detector(SSD) pipeline is adapted with DenseNet.In the experiments,the real pallets dataset is used for training and testing.In the testing phase,multi-scale feature maps of different resolutions are utilized for enhancing the network’s ability to adapt to detected objects,and the detection task is employing with single network.The experiment results show that the accuracy of the improved algorithm increased by 6.1% compared with YOLO algorithm.

Key words: object detection, pallet detection, Convolution Neural Betwork(CNN), deep learning, densely connected convolutional neural network

中图分类号:

TP391.4

李天剑,黄斌,刘江玉,金秋. 卷积神经网络物体检测算法在物流仓库中的应用[J]. 计算机工程, 2018, 44(6): 176-181.

LI Tianjian,HUANG Bin,LIU Jiangyu,JIN Qiu. Application of Convolution Neural Network Object Detection Algorithm in Logistics Warehouse[J]. Computer Engineering, 2018, 44(6): 176-181.

http://www.ecice06.com/CN/Y2018/V44/I6/176

参考文献

［1］LOWE D G.Distinctive image features from scale-invariant keypoints［J］.International Journal of Computer Vision,2004,60(2):91-110.
［2］SUYKENS J A K,VANDEWALLE J.Least squares support vector machine classifiers［J］.Neural Processing Letters,1999,9(3):293-300.
［3］DALAL N,TRIGGS B.Histograms of oriented gradients for human detection［C］//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2005:886-893.
［4］何希平,张琼华,刘波.基于HOG的目标分类特征深度学习模型［J］.计算机工程,2016,42(12):176-180.
［5］FELZENSZWALB P F,GIRSHICK R B,MCALLESTER D,et al.Object detection with discriminatively trained part-based models［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(9):1627-1645.
［6］REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:unified,real-time object detection［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:779-788.
［7］HAN J,MORAGA C.The influence of the sigmoid function parameters on the speed of backpropagation learning［C］//Proceedings of International Workshop on Artificial Neural Networks:From Natural to Artificial Neural Computation.Berlin,Germany:Springer-Verlag,1995:195-201.
［8］张蕾,章毅.大数据分析的无限深度神经网络方法［J］.计算机研究与发展,2016,53(1):68-79.
［9］KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenet classification with deep convolutional neural networks advances in neural information processing systems［C］//Proceedings of International Conference on Neural Infor-mation Processing System.Washington D.C.,USA:MIT Press,2012:1097-1105.
［10］GRAHAM B.Fractional max-pooling［EB/OL］.［2017-03-20］.https://arxiv.org/abs/1412.6071.
［11］SRIVASTAVA N,HINTON G E,KRIZHEVSKY A,et al.Dropout:a simple way to prevent neural networks from overfitting［J］.Journal of Machine Learning Research,2014,15(1):1929-1958.
［12］SRIVASTAVA N,HINTON G,KRIZHEVSKY A,et al.Dropout:a simple way to prevent neural networks from overfitting［J］.Journal of Machine Learning Research,2014,15(1):1929-1958.
［13］HUANG Gao,LIU Zhuang,LAURENS V D M,et al.Weinberger,densely connected convolutional networks［EB/OL］.［2017-03-20］.https://arxiv.org/abs/1608.06993v3.
［14］SRIVASTAVA R K,GREFF K,Schmidhuber J.Highway Networks［EB/OL］.［2017-03-20］.https://arxiv.org/abs/150 5.00387.
［15］HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:770-778.
［16］SZEGEDY C,LIU Wei,JIA Yangqing,et al.Going deeper with convolutions［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:1-9.
［17］LIU Wei,ANGUELOV D,ERHAN D,et al.SSD:single shot multibox detector［C］//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer-Verlag,2016:21-37.
［18］SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition［EB/OL］.［2017-03-20］.http://xueshu.baidu.com/s?wd=paperuri%3A%282801f41808e377a1897a3887b6758c59%29&filter=sc_long_sign&tn=SE_xueshusource_2kduw22v≻_vurl=http%3A%2F%2Fadsabs.harvard.edu%2Fabs%2F2014arXiv1409.1556S&ie=utf-8≻_us=4549459500002107255.
［19］SERMANET P,EIGEN D,ZHANG Xiang,et al.Overfeat:integrated recognition,localization and detection using convolutional networks［EB/OL］.［2017-03-20］.https://arxiv.org/abs/1312.6229v4.
［20］ERHAN D,SZEGEDY C,TOSHEV A,et al.Scalable object detection using deep neural networks［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2014:2147-2154.
［21］REN Shaoqing,HE Kaiming,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks［C］//Proceedings of the 28th International Conference on Neural Information Processing Systems.Montreal,Canada:MIT Press,2015:91-99.

[1]	江雨燕, 陶承凤, 李平. 数据增强和自适应自步学习的深度子空间聚类算法[J]. 计算机工程, 2023, 49(8): 96-103, 110.
[2]	李泽水, 冀俊忠, 杨翠翠. 基于边权重信息深度网络嵌入的PPIN功能模块检测[J]. 计算机工程, 2023, 49(8): 69-76.
[3]	王可铮, 徐玉芬, 周尚波. 结合对比感知损失和融合注意力的图像去雾模型[J]. 计算机工程, 2023, 49(8): 207-214.
[4]	刘俊豪, 王美林, 谢兴, 宋烨兴, 许莉花. 基于改进YOLOv5的皮革瑕疵检测算法[J]. 计算机工程, 2023, 49(8): 240-249.
[5]	闫兴亚, 匡娅茜, 白光睿, 李月. 基于深度学习的学生课堂行为识别方法[J]. 计算机工程, 2023, 49(7): 251-258.
[6]	李军侠, 王星驰, 殷梓, 石德硕. 边缘深度挖掘的弱监督显著性目标检测[J]. 计算机工程, 2023, 49(7): 169-178.
[7]	吴珊, 周凤. 基于改进SSD算法的小目标检测[J]. 计算机工程, 2023, 49(7): 179-188.
[8]	席建锐, 唐红梅, 梁春阳, 刘鑫. 基于改进隐函数的点云物体重建[J]. 计算机工程, 2023, 49(7): 214-222.
[9]	齐咏生, 杜晓旭, 朱俊峰, 高胜利, 刘利强. 基于增强型轻量深度网络的牧区牲畜高效检测[J]. 计算机工程, 2023, 49(7): 278-287.
[10]	谌雨章, 黄逸姿, 张钧涵. 基于多速率空洞卷积的多尺度水下小目标检测[J]. 计算机工程, 2023, 49(6): 257-264.
[11]	张博旭, 蒲智, 程曦. 基于提示学习的维吾尔语文本分类研究[J]. 计算机工程, 2023, 49(6): 292-299,313.
[12]	于海洋, 景鹏, 张文涛, 谢赛飞, 滑志华, 宋草原. 基于残差与注意力机制的道路裂缝检测U-Net改进模型[J]. 计算机工程, 2023, 49(6): 265-273.
[13]	王爱玲, 马文臻, 邹自明, 钟佳. 基于领域自适应的卫星工程参数异常检测[J]. 计算机工程, 2023, 49(5): 29-37,47.
[14]	宋羽凯, 谢江. 基于多任务学习的轻量级语音情感识别模型[J]. 计算机工程, 2023, 49(5): 122-128.
[15]	唐彦, 卢镘旭. 基于知识图谱与深度涟漪网络的推荐系统[J]. 计算机工程, 2023, 49(5): 63-72,80.

选择文件类型/文献管理软件名称

选择包含的内容

卷积神经网络物体检测算法在物流仓库中的应用

Application of Convolution Neural Network Object Detection Algorithm in Logistics Warehouse

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

卷积神经网络物体检测算法在物流仓库中的应用

Application of Convolution Neural Network Object Detection Algorithm in Logistics Warehouse

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价