登录    |    注册

您好,欢迎来到中国测试科技资讯平台!

在线投稿 专家审稿 编辑办公 咨询评价
首页> 《中国测试》期刊 >本期导读>面向智能维护的通信机房机柜图像语义分割技术

面向智能维护的通信机房机柜图像语义分割技术

44    2019-11-28

0.00

全文售价

作者:赵迪, 刘桂雄

作者单位:华南理工大学机械与汽车工程学院, 广东 广州 510640


关键词:智能维护;神经网络;语义分割;Mask R-CNN


摘要:

通信机房机柜的智能维护是实现设备无人化、智能化监管的核心工作之一,结合语义分割技术实现设备图像识别、位置检测、检修操作点确定,形成泛用性强的人工智能方法。该文从深度学习语义分割方法入手,提出基于Mask R-CNN的机房机柜设备图像语义分割技术方案,实现不同视野、存在物体遮挡条件下的机房机柜图像识别与分割。通过模拟不同语义分割算法在通信机房机柜检测场景的应用效果,表明基于Mask R-CNN的语义分割技术准确性良好,Top-1错误率为7.1%、像素级分割准确性mIOU达82.3%。


Image semantic segmentation method for cabinets of carrier hotel for intelligent maintenance
ZHAO Di, LIU Guixiong
School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640,China
Abstract: The intelligent maintenance of the cabinet in carrier hotel is the core of unmanned and intelligent supervision of the equipment. Combined with semantic segmentation technology, the device image recognition, position detection and maintenance operation point are determined, and a versatile artificial intelligence method is formed. This paper starts with the deep learning semantic segmentation method, and proposes the image semantic segmentation technology scheme of the cabinet equipment based on Mask R-CNN, which realizes the image recognition and segmentation of the cabinet under different vision and object occlusion conditions. By simulating the application effect of different semantic segmentation algorithms in the carrier hotel detection scenario, it is shown that the semantic segmentation technology based on Mask R-CNN has good accuracy, the Top-1 error rate is 7.1%, and the pixel-level segmentation accuracy mIOU reaches 82.3%.
Keywords: intelligent maintenance;neural network;semantic segmentation;Mask R-CNN
2019, 45(11):126-130  收稿日期: 2019-08-30;收到修改稿日期: 2019-09-28
基金项目: 广州市产业技术重大攻关计划(201802030006)
作者简介: 赵迪(1996-),男,江西九江市人,硕士研究生,专业方向为机器视觉技术研究
参考文献
[1] 李洋, 邓嘉, 张新艳. 串行伯努利生产线的实时维护决策控制[J]. 同济大学学报(自然科学版), 2018, 46(10):1410-1415
[2] 吴维农, 唐夲, 肖静薇, 等. 电力信息通信机房智能PDU监测系统的研究与实现[J]. 机床与液压, 2017, 45(6):83-88
[3] 舒送, 赵桂芳, 张小辉. 机器人技术在航空维修中的应用探索[J]. 机械设计与制造, 2018(1):248-250
[4] 宋屹峰, 凌烈, 姜勇, 等. 面向断股补修任务的输电线维护机器人设计[J]. 机械设计与制造, 2017(7):140-143
[5] 刘冬雨, 刘宏, 何宇, 等. 空间机械臂在轨维修的视觉伺服操控策略[J]. 机器人, 2018, 40(5):742-749
[6] 黄坚, 刘桂雄. 面向机器视觉检测的CNN语义分割方法进展[J]. 激光杂志, 2019, 40(5):10-16
[7] 刘桂雄, 黄坚, 刘思洋, 等. 面向语义分割机器视觉的AutoML方法[J]. 激光杂志, 2019, 40(6):1-9
[8] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651
[9] LIU G, HE B, LIU S, et al. Chassis assembly detection and identification based on deep learning component instance segmentation[J]. Symmetry, 2019, 11(6):1001
[10] 李良福, 孙瑞赟. 复杂背景下基于图像处理的桥梁裂缝检测算法[J]. 激光与光电子学进展, 2019, 56(6):112-122
[11] 朱曦阳, 张春, 张德兵. 智能机房监控系统中损失函数与图像合成的优化[J]. 微电子学与计算机, 2019, 36(1):37-40+45
[12] XIE S, TU Z. Holistically-nested edge detection[J]. International Journal of Computer Vision, 2015, 125(1-3):3-18
[13] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab:semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2018, 40(4):834-848
[14] HE K, GKIOXARI G, DOLLÁR P, et al. Mask R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision, 2017.