基于改进Faster R-CNN模型的水面漂浮物检测方法
River Drifting Garbage Detection Based on the Improved Faster R-CNN
DOI: 10.12677/CSA.2021.1112314, PDF,    科研立项经费支持
作者: 黄海源, 赵子豪*, 张海刚, 薛元飞：深圳职业技术学院粤港澳大湾区人工智能应用技术研究院，广东 深圳
关键词: 水面漂浮物；目标检测；卷积神经网络；Faster R-CNN；River Drifting Garbage； Object Detection； Convolutional Neural Network； Faster R-CNN
摘要: 水面环境治理长期以来都是生态环保的一项重要工作。然而，人工清理水面垃圾的方式难以满足实际工作需求。近年来，深度学习驱动的目标检测算法被成功应用在诸多领域。为解决传统检测方法效率低的问题，本文提出了一种水面漂浮物检测方法。该方法以Faster R-CNN模型为基础，对其主干网络做了改进。在自建的小型水面漂浮物数据集上进行实验，模型识别精度达到77.02%，相较于其它模型有至少2.56%的提升。此外，大量对比实验表明，该模型具有良好的检测性能，基本满足实际需求。
Abstract: River governance has always been a significant task of ecological environmental protection. However, manual cleaning for river drifting garbage is difficult to meet the actual work needs. In recent years, object detection algorithms have been successfully applied in many fields. To solve the problem of low efficiency of traditional methods, this paper proposes a detection method for river drifting garbage. This method is based on the Faster R-CNN model and improves its backbone network. The experiments are performed on the self-built small drifting garbage data set, and the model recognition accuracy reaches 77.02%, which is 2.56% higher than other classical models. Finally, some comparison experiments show that the proposed model has good detection performance and basically meets the actual needs.
文章引用：黄海源, 赵子豪, 张海刚, 薛元飞. 基于改进Faster R-CNN模型的水面漂浮物检测方法[J]. 计算机科学与应用, 2021, 11(12): 3108-3116. https://doi.org/10.12677/CSA.2021.1112314

参考文献

[1] 原建洋, 窦岩, 王少文, 刘赛赛. 市面上典型水面垃圾清理船对比研究[J]. 科技创新导报, 2019, 16(7): 67-68.
[2] Gu, J., Wang, Z., Kuen, J., Ma, L. and Wang, J. (2018) Recent Advances in Convolutional Neural Networks. Pattern Recognition, 77, 354-377. [Google Scholar] [CrossRef] 
[3] 曹诗雨, 刘跃虎, 李辛昭. 基于Fast R-CNN的车辆目标检测[J]. 中国图像图形学报, 2017, 22(5): 671-677.
[4] 刘聪聪, 应捷, 杨海马, 刘瑾, 李筠. 基于区域卷积神经网络的空中飞行物识别算法[J]. 传感器与微系统, 2021, 40(1): 110-113+117.
[5] 雷李义. 基于深度学习的水面漂浮物目标检测及分析[D]: [硕士学位论文]. 南宁: 广西大学, 2019.
[6] Dai, J., Li, Y., He, K. and Sun, J. (2016) Object Detection via Region-Based Fully Convolutional Networks. Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems 2016, Barcelona, 5-10 December 2016, 379-387.
[7] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S.E., Fu, C.Y. and Berg, A.C. (2016) Ssd: Single Shot Multibox Detector. European Conference on Computer Vision, Amsterdam, Vol. 1, 21-37. [Google Scholar] [CrossRef] 
[8] Simonyan, K. and Zisserman, A. (2015) Very Deep Convolu-tional Networks for Large-Scale Image Recognition. International Conference on Learning Representations, San Diego, 7-9 May 2015, 1-14.
[9] Redmon, J. and Farhadi, A. (2018) Yolov3: An Incremental Improvement.
[10] Redmon, J., Divvala, S.K., Girshick, R.B. and Farhadi, A. (2016) You Only Look Once: Unified, Real-Time Object Detection. IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 779-788. [Google Scholar] [CrossRef] 
[11] Redmon, J. and Farhadi, A. (2017) YOLO9000: Better, Faster, Stronger. IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 6517-6525. [Google Scholar] [CrossRef] 
[12] Ioffe, S. and Szegedy, C. (2015) Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. International Conference on Machine Learning, Lille, 7-9 July 2015, 448-456.
[13] Bochkovskiy, A., Wang, C.Y. and Liao, H.Y.M. (2020) Yolov4: Optimal Speed and Accura-cy of Object Detection.
[14] He, K., Zhang, X., Ren, S. and Sun, J. (2015) Spatial Pyramid Pooling in Deep Convolu-tional Networks for Visual Recognition. IEEE Transactions on Pattern Analysis & Machine Intelligence, 37, 1904-1916. [Google Scholar] [CrossRef] 
[15] Ren, S., He, K., Girshick, R.B. and Sun, J. (2017) Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis & Machine Intelligence, 39, 1137-1149. [Google Scholar] [CrossRef] 
[16] Girshick, R.B., Donahue, J., Darrell, T. and Malik, J. (2014) Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. IEEE Conference on Computer Vi-sion and Pattern Recognition, Columbus, 23-28 June 2014, 580-587. [Google Scholar] [CrossRef] 
[17] Girshick R.B. (2015) Fast R-CNN. IEEE International Conference on Computer Vision, Santiago, 7-13 December 2015, 1440-1448. [Google Scholar] [CrossRef] 
[18] He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep Residual Learning for Image Recognition. IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 770-778. [Google Scholar] [CrossRef]