基于HMM的国网轮询视频分会场名称识别

doi:10.6040/j.issn.1672-3961.0.2021.359

山东大学学报 (工学版) ›› 2022, Vol. 52 ›› Issue (6): 183-190.doi: 10.6040/j.issn.1672-3961.0.2021.359

• 电气工程 • 上一篇

基于HMM的国网轮询视频分会场名称识别

何子亨¹,孙丽丽¹,左修洋²,刘鸿雁¹,王雨晨¹,车四四¹,王朔¹

1. 国网山东省电力公司信息通信公司, 山东济南 250001;2. 山东大学信息科学与工程学院, 山东青岛 266000

发布日期:2022-12-23
作者简介:何子亨(1991— ),男,山东烟台人,工程师,硕士研究生,主要研究方向为电力系统通信. E-mail: 413501559@qq.com
基金资助:
国网山东省电力公司科技资助项目(520627210004)

Branch venue name recognition for State Grid polling video based on HMM

HE Ziheng¹, SUN Lili¹, ZUO Xiuyang², LIU Hongyan¹, WANG Yuchen¹, CHE Sisi¹, WANG Shuo¹

1. Information &
Telecommunications Company, State Grid Shandong Electric Power Company, Jinan 250001, Shandong, China;
2. School of Information Science and Engineering, Shandong University, Qingdao 266000, Shandong, China

Published:2022-12-23

摘要/Abstract

摘要： 为提高运维效率,针对视频中的分会场文字信息,采用计算机视觉技术,识别出分会场的名称,以便实现轮询视频的自动检测。提出一种基于隐马尔可夫模型(hidden Markov model,HMM)的轮询视频分会场名称识别算法,利用分会场名称中相邻单个文字的相关性,提高分会场名称的识别准确率。对每帧视频图像,采用微分二值化(differentiable binarization,DB)算法定位文字区域,提取单个文字的分块特征,并通过计算欧式距离进行单字识别。考虑分会场名称中相邻文字之间的相关性,构建HMM,实现相邻文字之间的关联,并采用Viterbi算法计算分会场名称识别结果。试验数据表明,在采用较低维数的特征向量时,本研究提出的分会场名称识别算法具有较高的识别率和较强的抗噪性能。

关键词: 文字识别, 视频会议, DB算法, HMM, Viterbi算法, 国家电网, 计算机视觉

中图分类号:

TP391

何子亨,孙丽丽,左修洋,刘鸿雁,王雨晨,车四四,王朔. 基于HMM的国网轮询视频分会场名称识别[J]. 山东大学学报 (工学版), 2022, 52(6): 183-190.

HE Ziheng, SUN Lili, ZUO Xiuyang, LIU Hongyan, WANG Yuchen, CHE Sisi, WANG Shuo. Branch venue name recognition for State Grid polling video based on HMM[J]. Journal of Shandong University(Engineering Science), 2022, 52(6): 183-190.

参考文献

[1] JUNG-PIL Shin. Optimal stroke-correspondence search method for on-line character recognition[J]. Pattern Recognition Letters, 2002, 23(5): 601-608.
[2] RADWAN M A, KHALIL M I, ABBAS H M. Neural networks pipeline for offline machine printed arabic OCR[J]. Neural Processing Letters, 2018, 48: 769-787.
[3] PANAHI R, GHOLAMPOUR I. Accurate detection and recognition of dirty vehicle plate numbers for high-speed applications[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(4): 767-779.
[4] LIAO Minghui, WAN Zhaoyi, YAO Cong, et al. Real-time scene text detection with differentiable binarization[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7): 11474-11481.
[5] LIAO Minghui, LYU Pengyuan, HE Minghang, et al. Mask textspotter: an end-to-end trainable neural network for spotting text with arbitrary shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021,43(2): 532-548.
[6] HE Tong, HUANG Weilin, QIAO Yu, et al. Text-attentional convolutional neural network for scene text detection[J]. IEEE Transactions on Image Processing, 2016, 25(6): 2529-2541.
[7] LIAO Minghui, SHI Baoguang, BAI Xiang. Textboxes++: a single-shot oriented scene text detector[J]. IEEE Transactions on Image Processing, 2018, 27(8): 3676-3690.
[8] ALOTAIBI Y A, SELOUANI S A, YAKOUB M S, et al. A canonicalization of distinctive phonetic features to improve arabic speech recognition[J]. Acta Acustica United with Acustica, 2019, 105(6): 1269-1277.
[9] LEE S, KHAN M Q, HUSEN M N. Continuous car driving intent detection using structural pattern recognition[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 99: 1-13.
[10] HOU Haijing, JIN Lisheng, NIU Qingning, et al. Driver intention recognition method using continuous hidden Markov model[J]. International Journal of Computational Intelligence Systems, 2011, 4(3): 386-393.
[11] FU Rongrong, WANG Hong, ZHAO Wenbo. Dynamic driver fatigue detection using hidden Markov model in real driving condition[J]. Expert Systems with Applications, 2016, 63: 397-411.
[12] SONG Chengbo, YAN Xuefeng, STEPHEN Nkyi, et al. Hidden Markov model and driver path preference for floating car trajectory map matching[J]. IET Intelligent Transport Systems, 2018, 12(10): 1433-1441.
[13] 陈炳鑫, 陈黎飞. 符号序列的预训练HMM分类方法[J]. 南京大学学报(自然科学), 2021, 57(1): 52-58. CHEN Bingxin, CHEN Lifei. A pretraining HMM classification method for symbolic sequences[J]. Journal of Nanjing University(Natural Science), 2021, 57(1): 52-58.
[14] CHEN Jiansheng, KANG Xiangui, LIU Ye, et al. Median filtering forensics based on convolutional neural networks[J]. IEEE Signal Processing Letters, 2015, 22(11): 1849-1853.
[15] DAVID G Lowe. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110.
[16] 牛畅,黄银和,尹奎英. 基于分块SURF特征提取的图像目标跟踪算法[J]. 激光与红外, 2017, 47(12): 1541-1547. NIU Chang, HUANG Yinhe, YIN Kuiying. Image target tracking algorithm based on blocked SURF extraction[J]. Laser & Infrared, 2017, 47(12): 1541-1547.
[17] MA Guohong, WANG Cong, LIU Pei, et al. Sequential similarity detection algorithm based on image edge feature[J]. Journal of Shanghai Jiaotong University(Science), 2014, 19(1): 79-83.
[18] 刘彩霞,卢干强,汤红波,等. 一种基于Viterbi算法的虚拟网络功能自适应部署方法[J]. 电子与信息学报, 2016, 38(11): 2922-2930. LIU Caixia, LU Ganqiang, TANG Hongbo, et al. Adaptive deployment method for virtualized network function based on Viterbi algorithm[J]. Journal of Electronics & Information Technology, 2016, 38(11): 2922-2930.
[19] 李丽敏,王仲生,姜洪开. 基于多状态MOG-HMM和Viterbi的航空发动机突发故障预测[J]. 振动、测试与诊断, 2014, 34(2): 310-314. LI Limin, WANG Zhongsheng, JIANG Hongkai. Aero-engine abrupt failure prognosis based on multi-states MOG-HMM and Viterbi algorithm[J]. Journal of Vibration, Measurement & Diagnosis, 2014, 34(2): 310-314.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于HMM的国网轮询视频分会场名称识别

Branch venue name recognition for State Grid polling video based on HMM

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

多维度评价

本文评价

推荐阅读 0

[1]	赵子健,陈兆瑞,李冰清. 基于非最小化优化的手眼标定方法[J]. 山东大学学报(工学版), 2016, 46(4): 28-33.
[2]	丛奎荣韩杰常发亮. 视觉机器人货物轮廓提取与定位[J]. 山东大学学报(工学版), 2010, 40(1): 15-18.