改进的Faster R-CNN方法及其在电缆隧道积水定位识别中的应用
|
摘要
针对电缆隧道内积水的问题,提出了一种改进的基于区域建议的卷积神经网络(Faster R-CNN)方法,并将其应用在电缆隧道积水定位识别中。考虑到Softmax分类方法的正则化参数选取会引起概率计算产生问题,改用支持向量机(SVM)进行图像分类,以增强分类的置信度。使用区域建议网络(RPN)提取隧道积水原图中的区域建议,然后用Fast R-CNN检测网络在建议框中进行图像识别、SVM分类和位置精修。实验结果表明,所提方法计算速度快、识别精度高,在实际工程中表现出较高的效率。
Aiming at the water accumulation problem in cable tunnel,an improved Faster R-CNN(Faster Region-based Convolutional Neural Network) method is proposed and applied in the recognition of cable tunnel water accumulation. Considering that the regularization parameter selection of Softmax may have problems with calculating the probability,the SVM(Support Vector Machine) is used to classify images to improve the classification accuracy. The RPN(Region Proposal Network) is used to extract proposals from the original images of cable tunnel water accumulation,and then the detection network of Fast R-CNN is used to carry out image recognition,SVM classification and location refining. The experiment results show that the proposed algorithm has the advantages of fast calculation speed,high recognition accuracy and high efficiency in practical engineering.
Aiming at the water accumulation problem in cable tunnel,an improved Faster R-CNN(Faster Region-based Convolutional Neural Network) method is proposed and applied in the recognition of cable tunnel water accumulation. Considering that the regularization parameter selection of Softmax may have problems with calculating the probability,the SVM(Support Vector Machine) is used to classify images to improve the classification accuracy. The RPN(Region Proposal Network) is used to extract proposals from the original images of cable tunnel water accumulation,and then the detection network of Fast R-CNN is used to carry out image recognition,SVM classification and location refining. The experiment results show that the proposed algorithm has the advantages of fast calculation speed,high recognition accuracy and high efficiency in practical engineering.
引文
[1] 费斐.电力隧道监控系统设计[J].科技信息:科学教研,2007(23):386,373.FEI Fei.The design of electric tunnel monitoring system[J].Science and Technology Information:Science Teaching and Research,2007(23):386,373.
[2] 朱丽梅.基于PLC的电缆隧道无线分布式监控系统[J].电力自动化设备,2004,24(9):71-73.ZHU Limei.PLC-based cable tunnel wireless distributed monitoring system[J].Electric Power Automation Equipment,2004,24(9):71-73.
[3] CUN Y L,BOSER B,DENKER J S,et al.Handwritten digit recognition with a back-propagation network[C]∥Advances in Neural Information Processing Systems.[S.l.]:Morgan Kaufmann Publishers Inc.,1990:396-404.
[4] HE Kaimin,ZHANG Xiangyu,REN Shaoqing,et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1906.
[5] HINTON G,LI Deng,YU Dong,et al.Deep neural networks for acoustic modeling in speech recognition:the shared views of four research groups[J].IEEE Signal Processing Magazine,2012,29(6):82-97.
[6] GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hie-rarchies for accurate object detection and semantic segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition.Columbus,OH,USA:IEEE,2014:580-587.
[7] LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Boston,MA,USA:IEEE,2015:3431-3440.
[8] ZITNICK C L,DOLLáR P.Edge Boxes:locating object proposals from edges[C]//European Conference on Computer Vision.Cham,Switzerland:Springer,2014:391-405.
[9] REN Shaoqing,HE Kaiming,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
[10] 桑军,郭沛,项志立,等.Faster-RCNN车型识别分析[J].重庆大学学报,2017,40(7):32-36.SANG Jun,GUO Pei,XIANG Zhili,et al.The analysis of the model identification of Faster-RCNN[J].Journal of Chongqing University,2017,40(7):32-36.
[11] 王万国,田兵,刘越,等.基于RCNN的无人机巡检图像电力小部件识别研究[J].地球信息科学学报,2017,19(2):256-263.WANG Wanguo,TIAN Bing,LIU Yue,et al.Research on the identification of UAV patrol image based on RCNN[J].Journal of Earth Information Science,2017,19(2):256-263.
[12] 张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.ZHANG Xuegong.On statistical learning theory and support vector machine[J].Journal of Automation,2000,26(1):32-42.
[13] 张哲,赵文清,朱永利,等.基于支持向量回归的电力变压器状态评估[J].电力自动化设备,2010,30(4):81-84.ZHANG Zhe,ZHAO Wenqing,ZHU Yongli,et al.State evaluation of power transformers based on support vector regression[J].Electric Power Automation Equipment,2010,30(4):81-84.
[14] 张镱议,焦健,汪可,等.基于帝国殖民竞争算法优化支持向量机的电力变压器故障诊断模型[J].电力自动化设备,2018,38(1):99-104.ZHANG Yiyi,JIAO Jian,WANG Ke,et al.Power transformer fault diagnosis model based on imperial colonial competition algorithm optimized support vector machine[J].Electric Power Automation Equipment,2018,38(1):99-104.
[15] 郭明玮,赵宇宙,项俊平,等.基于支持向量机的目标检测算法综述[J].控制与决策,2014,29(2):193-200.GUO Mingwei,ZHAO Yuzhou,XIANG Junping,et al.Overview of target detection algorithm based on support vector machine[J].Control and Decision,2014,29(2):193-200.
[2] 朱丽梅.基于PLC的电缆隧道无线分布式监控系统[J].电力自动化设备,2004,24(9):71-73.ZHU Limei.PLC-based cable tunnel wireless distributed monitoring system[J].Electric Power Automation Equipment,2004,24(9):71-73.
[3] CUN Y L,BOSER B,DENKER J S,et al.Handwritten digit recognition with a back-propagation network[C]∥Advances in Neural Information Processing Systems.[S.l.]:Morgan Kaufmann Publishers Inc.,1990:396-404.
[4] HE Kaimin,ZHANG Xiangyu,REN Shaoqing,et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1906.
[5] HINTON G,LI Deng,YU Dong,et al.Deep neural networks for acoustic modeling in speech recognition:the shared views of four research groups[J].IEEE Signal Processing Magazine,2012,29(6):82-97.
[6] GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hie-rarchies for accurate object detection and semantic segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition.Columbus,OH,USA:IEEE,2014:580-587.
[7] LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Boston,MA,USA:IEEE,2015:3431-3440.
[8] ZITNICK C L,DOLLáR P.Edge Boxes:locating object proposals from edges[C]//European Conference on Computer Vision.Cham,Switzerland:Springer,2014:391-405.
[9] REN Shaoqing,HE Kaiming,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
[10] 桑军,郭沛,项志立,等.Faster-RCNN车型识别分析[J].重庆大学学报,2017,40(7):32-36.SANG Jun,GUO Pei,XIANG Zhili,et al.The analysis of the model identification of Faster-RCNN[J].Journal of Chongqing University,2017,40(7):32-36.
[11] 王万国,田兵,刘越,等.基于RCNN的无人机巡检图像电力小部件识别研究[J].地球信息科学学报,2017,19(2):256-263.WANG Wanguo,TIAN Bing,LIU Yue,et al.Research on the identification of UAV patrol image based on RCNN[J].Journal of Earth Information Science,2017,19(2):256-263.
[12] 张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.ZHANG Xuegong.On statistical learning theory and support vector machine[J].Journal of Automation,2000,26(1):32-42.
[13] 张哲,赵文清,朱永利,等.基于支持向量回归的电力变压器状态评估[J].电力自动化设备,2010,30(4):81-84.ZHANG Zhe,ZHAO Wenqing,ZHU Yongli,et al.State evaluation of power transformers based on support vector regression[J].Electric Power Automation Equipment,2010,30(4):81-84.
[14] 张镱议,焦健,汪可,等.基于帝国殖民竞争算法优化支持向量机的电力变压器故障诊断模型[J].电力自动化设备,2018,38(1):99-104.ZHANG Yiyi,JIAO Jian,WANG Ke,et al.Power transformer fault diagnosis model based on imperial colonial competition algorithm optimized support vector machine[J].Electric Power Automation Equipment,2018,38(1):99-104.
[15] 郭明玮,赵宇宙,项俊平,等.基于支持向量机的目标检测算法综述[J].控制与决策,2014,29(2):193-200.GUO Mingwei,ZHAO Yuzhou,XIANG Junping,et al.Overview of target detection algorithm based on support vector machine[J].Control and Decision,2014,29(2):193-200.