首页 > 论文 > 激光与光电子学进展 > 57卷 > 20期(pp:201014--1)

基于Faster RCNN的生活垃圾智能识别

Intelligent Domestic Garbage Recognition Based on Faster RCNN

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

利用Faster RCNN算法实现生活垃圾的高精度识别。选取典型的6种生活垃圾建立数据集,采用数据增强方法提升了数据集目标数量及目标类别、尺度均衡性,分析对比三种具有显著差异的主干网络VGG-16、Res101、MobileNet_v1的精度、速度及泛化性能。采用结合特异层微调的端到端训练策略,对低识别率样本开展增强训练,由此获得了最低为92.85%的均值平均精度(mAP),随后对误识别样本中提取的三种典型错误进行优化,将最高mAP提高到99.23%。此外,设计含816张图片的背景数据集测试算法在多变背景下的泛化性能,发现复杂垃圾背景对检测精度的影响最大,且泛化性能与网络收敛性能趋势一致,即从优到劣排序依次为Res101、 VGG-16、MobileNet_v1。最后,基于可回收垃圾倾向高精度指标及有害垃圾倾向高召回率指标的原则,分析并得到算法最优检测概率阈值的设置方法。

Abstract

In this paper, we presented the Intelligent Domestic Garbage Recognition using Faster RCNN to realize high-precision identification of domestic garbage. Specifically, 6 kinds of domestic garbage were selected to build the dataset. The data augmentation technique was adopted to expand the quantity and category of the targets, and improve balance on the size of the targets. Moreover, we used three different types of backbone networks including VGG-16, Res101, and MobileNet_v1 to analyze and compare the accuracy, speed, and generalization performance. The research used end-to-end training network finely tuned by the special layer, and carried out enhanced training on low recognition rate samples to obtain a minimum mean average precision (mAP) of 92.85%. Subsequently, we captured three typical errors and optimized from the misidentified samples, and thus the highest recognition mAP increased to 99.23%. To analyze the generalization performance of different backbone networks embedded in the algorithm, we built a dataset with 816 pictures derivatized from the different backgrounds and used it to test the impact of changing the background on garbage detection. As a result, we found that the complex backgrounds from surrounding garbage put the greatest impact on detection accuracy. Thus, the generalization performance takes the same trend as convergence performance, which changes Res101, VGG-16, MobileNet_v1 from good to bad. Therefore, the setting method of the optimal probability threshold for algorithm detection was analyzed and obtained based on the principles of the high-precision requirement for recyclable garbage and high recall requirements for hazardous garbage.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TP391.4; X705

DOI:10.3788/LOP57.201014

所属栏目:图像处理

收稿日期:2020-01-13

修改稿日期:2020-03-09

网络出版日期:2020-10-01

作者单位    点击查看

文灿华:上海交通大学中英国际低碳学院, 上海 201306
李佳:上海交通大学中英国际低碳学院, 上海 201306
董雪:上海交通大学中英国际低碳学院, 上海 201306

联系人作者:李佳(canhuamail@163.com)

【1】Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]∥2014 IEEE Conference on Computer Vision and Pattern Recognition, June 23-28, 2014, Columbus, OH, USA. New York: , 2014, 580-587.

【2】Redmon J, Divvala S, Girshick R, et al. You only look once: , 2016, 779-788.

【3】Redmon J, Farhadi A. YOLO9000: , 2017, 6517-6525.

【4】Redmon J. -04-08)[2020-01-03] . https: ∥arxiv. 2018, org/abs/1804: 02767.

【5】Liu W, Anguelov D, Erhan D, et al. SSD: single shot multibox detector . [C]∥European Conference on Computer Vision. Cham: Springer. 2016, 21-37.

【6】Ren S Q, He K M, Girshick R, et al. -06-04)[2020-01-03] . https: ∥arxiv. 2015, org/abs/1506: 01497.

【7】Chen Z H, Zou H B, Wang Y B, et al. A vision-based robotic grasping system using deep learning for garbage sorting[C]∥2017 36th Chinese Control Conference (CCC), July 26-28, 2017, Dalian, China. New York: , 2017, 11223-11226.

【8】Mikami K, Chen Y, Nakazawa J, et al. DeepCounter: , 2018, 1-10.

【9】Zhang P C, Zhao Q, Gao J, et al. Urban street cleanliness assessment using mobile edge computing and deep learning [J]. IEEE Access. 2019, 7: 63550-63563.

【10】Everingham M, van Gool L, Williams C K I, et al. The pascal visual object classes (VOC) challenge [J]. International Journal of Computer Vision. 2010, 88(2): 303-338.

【11】He K M, Gkioxari G, Dollár P, et al. Mask R-CNN[C]∥2017 IEEE International Conference on Computer Vision (ICCV), October 22-29, 2017, Venice, Italy. New York: , 2017, 2980-2988.

【12】Simonyan K. -09-04)[2020-01-03] . https: ∥arxiv. 2014, org/abs/1409: 1556.

【13】He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]∥2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 27-30, 2016, Las Vegas, NV, USA. New York: , 2016, 770-778.

【14】Howard A G, Zhu M L, Chen B, et al. -04-17)[2020-01-03] . https: ∥arxiv. 2017, org/abs/1704: 04861.

【15】Russakovsky O, Deng J, Su H, et al. ImageNet large scale visual recognition challenge [J]. International Journal of Computer Vision. 2015, 115(3): 211-252.

【16】Zeiler M D, Fergus R. Visualizing and understanding convolutional networks . [C]∥European Conference on Computer Vision. Cham: Springer. 2014, 818-833.

引用该论文

Wen Canhua,Li Jia,Dong Xue. Intelligent Domestic Garbage Recognition Based on Faster RCNN[J]. Laser & Optoelectronics Progress, 2020, 57(20): 201014

文灿华,李佳,董雪. 基于Faster RCNN的生活垃圾智能识别[J]. 激光与光电子学进展, 2020, 57(20): 201014

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF