改进YOLOX-s算法的自动贩卖机商品检测方法

doi:10.3778/j.issn.1002-8331.2301-0015

摘要/Abstract

摘要： 针对自动贩卖机商品检测中图片变形、遮挡及光线环境，导致各商品细粒度检测性能低问题，构建一种基于YOLOX-s网络改进的算法模型YOLOX-s-BGC。通过改进的双向特征金字塔网络（BiFPN-m），减小网络特征融合过程中特征信息的丢失，并提高了模型的推理速度；同时引入Ghost卷积降低参数量以减少网络计算开销；为了可以关注图像中更具区分度的特征信息，还提出了卷积块注意力模块（CBAM），提取出更具区分性的特征。在自动贩卖机商品检测数据集上的实验结果表明，YOLOX-s-BGC模型在商品检测的检测精度达到了99.57%，相比于原始YOLOX-s算法提高了1.91个百分点，且计算参数量和模型大小基本保持不变。同时与SSD、YOLOv3、Scaled YOLOv4、YOLOv5 Lite-g等目标检测算法相比，该改进算法具有一定的优越性，是在自动贩卖机商品检测中的理想模型。

关键词: 商品检测, 目标检测, YOLOX, 注意力机制, 轻量化网络

Abstract: In view of the low performance of fine-grained detection of goods caused by image deformation, occlusion and light environment in vending machine goods detection, an improved algorithm model YOLOX-s-BGC based on YOLOX-s network is constructed. The improved bi-directional feature pyramid network（BiFPN-m） reduces the loss of feature information in the process of network feature fusion, and improves the reasoning speed of the model. At the same time, ghost convolution is introduced to reduce the amount of parameters to reduce the network computing overhead. In order to focus on the more discriminative feature information in the image, the convolution block attention module（CBAM） is also proposed to extract more discriminative features. The experimental results on the vending machine commodity detection data set show that the detection accuracy of YOLOX-s-BGC model in commodity detection reaches 99.57%, which is 1.91 percentage points higher than the original YOLOX-s algorithm, and the calculation parameters and model size are basically unchanged. Comparing with SSD, YOLOv3, Scaled YOLOv4, YOLOv5 Lite-g and other target detection algorithms, the improved algorithm has certain advantages, and is an ideal model in vending machine commodity detection.

Key words: commodity detection, target detection, YOLOX, attention mechanism, lightweight network

张少林, 姜吴瑾, 李太福, 杨杰. 改进YOLOX-s算法的自动贩卖机商品检测方法[J]. 计算机工程与应用, 2023, 59(20): 274-282.

ZHANG Shaolin, JIANG Wujin, LI Taifu, YANG Jie. Commodity Detection Method for Vending Machines Based on Improved YOLOX-s Algorithm[J]. Computer Engineering and Applications, 2023, 59(20): 274-282.

参考文献

[1] YANG C，WANG X，MAO S.RFID tag localization with a sparse tag array[J].IEEE Internet of Things Journal，2021，9（18）：16976-16989.
[2] TONIONI A，DI STEFANO L.Domain invariant hierarchical embedding for grocery products recognition[J].Computer Vision and Image Understanding，2019，182：81-92.
[3] KRIZHEVSKY A，SUTSKEVER I，HINTON G E.Imagenet classification with deep convolutional neural networks[J].Communications of the ACM，2017，60（6）：84-90.
[4] GIRSHICK R，DONAHUE J，DARRELL T，et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition（CVPR），Columbus，Jun 23-28，2014.Washington DC：IEEE Computer Society，2014：580-587.
[5] GIRSHICK R.Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision（ICCV），Santiago，Dec 7-13，2015.Piscataway，NJ：IEEE，2015：1440-1448.
[6] REN S，HE K，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems，2015.
[7] REDMON J，FARHADI A.YOLOv3：an incremental improvement[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition（CVPR），Salt Lake City，June 18-22，2018.Piscataway，NJ：IEEE，2018：89-95.
[8] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.YOLOv4：optimal speed and accuracy of objectt detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition（CVPR），Seattle，Jun 14-19，2020.Piscataway，NJ：IEEE，2020.
[9] GE Z，LIU S，WANG F，et al.Yolox：exceeding YOLO series in 2021[EB/OL].（2021-08-06）[2022-12-25].https：//arxiv.org/abs/2107.084-30.
[10] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[C]//Proceedings of the 14th European Conference on Computer Vision（ECCV），Amsterdam，Oct 10-16，2016.Berlin：Springer，2016：21-37.
[11] HAN K，WANG Y，CHEN H，et al.A survey on vision transformer[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2022，45（1）：87-110.
[12] 刘照邦，袁明辉.基于深度神经网络的货架商品识别方法[J].包装工程，2020，41（1）：149-155.
LIU Z B，YUAN M H.Product recognition on shelves based on deep neural network[J].Packaging Engineering，2020，41（1）：149-155.
[13] JI Z，KONG Q，WANG H，et al.Small and dense commodity object detection with multi-scale receptive field attention[C]//Proceedings of the 27th ACM International Conference on Multimedia，Nice，Oct 21-25，2019.New York：ACM Press，2019：1349-1357.
[14] 刘文豪，姜胜明.基于无标签半监督学习的商品识别方法[J].计算机应用与软件，2022，39（7）：167-173.
LIU W H，JIANG S M.Product identification method based on unlabeled semi-supervised learning[J].Computer Applications，2022，39（7）：167-173.
[15] HAN K，WANG Y，TIAN Q，et al.Ghostnet：more features from cheap operations[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，Seattle，Jun 14-19，2020.Piscataway，NJ：IEEE，2020：1580-1589.
[16] WOO S，PARK J，LEE J Y，et al.CBAM：convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision（ECCV），Munich，Sep 8-14，2018.Berlin：Springer，2018：3-19.
[17] TAN M，PANG R，LE Q V.Efficientdet：scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），Seattle，Jun 14-19，2020.Piscataway，NJ：IEEE，2020：10781-10790.
[18] 刘剑峰，潘晨.增强特征金字塔结构的显著目标检测算法[J].计算机工程与应用，2022，58（12）：226-233.
LIU J F，PAN C.Salient object detection for enhanced feature pyramid structure[J].Computer Engineering and Applications，2022，58（12）：226-233.
[19] 钟志峰，夏一帆，周冬平，等.基于改进YOLOv4的轻量化目标检测算法[J].计算机应用，2022，42（7）：2201-2209.
ZHONG Z F，XIA Y F，ZHOU D P，et al.Lightweight object detection algorithm based on improved YOLOv4[J].Journal of Computer Applications，2022，42（7）：2201-2209.
[20] GU Y，QIN X，PENG Y，et al.Content-augmented feature pyramid network with light linear spatial transformers for object detection[J].IET Image Processing，2022，16（13）：3567-3578.
[21] 王玲敏，段军，辛立伟.引入注意力机制的YOLOv5安全帽佩戴检测方法[J].计算机工程与应用，2022，58（9）：303-312.
WANG L M，DUAN J，XIN L W.YOLOv5 helmet wear detection method with introduction of attention mechanism[J].Computer Engineering and Applications，2022，58（9）：303-312.
[22] WANG C Y，BOCHKOVSKIY A，LIAO H Y M.Scaled-yolov4：scaling cross stage partial network[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），Jun 19-25，2021.Piscataway，NJ：IEEE，2021：13029-13038.
[23] CHEN X，GONG Z.YOLOv5-Lite：lighter，faster and easier to deploy[EB/OL].（2021-04-24）[2022-12-25].https：//doi.org/10.5281/zenodo.5241425.