图像识别技术在电信运维质检上的应用

范博; 邱芸; 沈雷

doi:10.11959/j.issn.1000-0801.2019086

您当前的位置：

首页 >

文章列表页 >

图像识别技术在电信运维质检上的应用

运行商人工智能专栏 | 更新时间：2024-06-05

- 图像识别技术在电信运维质检上的应用
- Application of image detection techniques in telecom operation and maintenance system
- 电信科学 2019年35卷第4期页码：146-152
- 作者机构：
- 作者简介：
  
  [ "范博（1993- ），男，中国电信股份有限公司上海研究院数据挖掘工程师，主要研究方向为图像算法研究和数据挖掘。" ]
  [ "邱芸（1980- ），女，中国电信股份有限公司上海研究院网络系统架构师，主要研究方向为基于网络数据的应用研究开发、物联网网络数据分析。" ]
  [ "沈雷（1970- ），男，中国电信股份有限公司上海研究院算法专家，主要研究方向为移动网络应用研究开发、人工智能算法。" ]
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2019086
  中图分类号： TN929.5
- 网络出版日期：2019-04，
  
  纸质出版日期：2019-04-20
- 稿件说明：
移动端阅览
范博, 邱芸, 沈雷. 图像识别技术在电信运维质检上的应用[J]. 电信科学, 2019,35(4):146-152.

Bo FAN, Yun QIU, Lei SHEN. Application of image detection techniques in telecom operation and maintenance system[J]. Telecommunications science, 2019, 35(4): 146-152.
范博, 邱芸, 沈雷. 图像识别技术在电信运维质检上的应用[J]. 电信科学, 2019,35(4):146-152. DOI： 10.11959/j.issn.1000-0801.2019086.

Bo FAN, Yun QIU, Lei SHEN. Application of image detection techniques in telecom operation and maintenance system[J]. Telecommunications science, 2019, 35(4): 146-152. DOI： 10.11959/j.issn.1000-0801.2019086.

摘要

随着电信行业的发展，持续增长的宽带业务需要更高的人工运维成本。在电信运维的装维质检中

需要人工识别施工现场图片以评估装维质量。传统的装维人工质检存在检测准确率低且人力成本高的问题。近年来基于深度学习的图像识别技术不断发展，应用领域不断拓宽。若采用图像识别技术，则可显著节约人力成本并提高准确率。将多种图像识别技术应用在电信运维业务中的装维场景，并分析比较各种图像识别技术的性能与准确率。基于分析，进一步提出了一种融合的图像识别模型，应用于电信运维的质检流程，显著提高了运维图片识别准确率和质检效率。

Abstract

With the development of telecom industry

growing web service operations need more people to maintain the quality of the service.In the branch of telecom operation service

installation and maintenance service

the service quality is evaluated by human through assessing picture of construction site.The traditional way of evaluating quality of operation service brings low accuracy and high labor cost.In recent years

image recognition technology developed well with the deep learning raised

the related application field were broaden.If image recognition techniques were applied in telecom operation and maintenance system

the labor cost would be saved and the recognition accuracy would be improved.Several image recognition techniques were applied in installation quality assessment of telecom operation and maintenance system

and the performance and accuracy were analyzed.Based on analysis

a combining image recognition model was further proposed.The image recognition accuracy and efficiency of quality assessment was improved.

关键词

Keywords

references

ANDREOPOULOS A , TSOTSOS J K . 50 years of object recognition:directions forword [J ] . Computer Vision and Image Understanding , 2013 , 117 ( 8 ):82789.

RUSSAKOVSKY O , DENG J , SU H , et al . ImageNet:large scal visual recognition challenge [J ] . International Journal of Computer Vision , 2015 , 115 ( 3 ): 211 - 252 .

KRIZHEVSKY A , SUTSKEVER I , HINTONG E . ImageNet classification with deep convolutional neural networks [C ] // The 25th International Conference on Neural Information Processing Systems,Dec 3-6,2012,Lake Tahoe,Nevada,USA . New York:ACM Press , 2012 : 1097 - 1105 .

HE K , ZHANG X Y , RENS Q , et al . Spatial pyramid poling in deep convolutional networks for visual recognition [J ] . IEEE Trans on Pattern Analysis and Machine Intelligence , 2015 , 37 ( 9 ): 1904 - 1914 .

HE K , GKIOXARI G , DOLLAR P , et al . Mask R-CNN [C ] // 2017 IEEE International Conference on Computer Vision,Oct 22-29,2017,Venice,Italy . Piscataway:IEEE Press , 2017 : 2980 - 2988 .

浏览量

431

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

多目标滚动时域优化的数学模型在运营商智能派单中的构建与应用

电信运营商AI客服平台技术研究与应用分析

5G车联网资源优化分配方案综述

运营商智能客服的关键技术和发展趋势

基于网络演进的人工智能技术方向研究