结合全局和局部约束的sLDA铁路扣件分类模型

doi:10.11772/j.issn.1001-9081.2018081767

计算机应用 ›› 2019, Vol. 39 ›› Issue (3): 888-893.DOI: 10.11772/j.issn.1001-9081.2018081767

• 虚拟现实与多媒体计算 • 上一篇下一篇

结合全局和局部约束的sLDA铁路扣件分类模型

杨飞, 罗建桥, 李柏林

西南交通大学机械工程学院, 成都 610031

收稿日期:2018-08-24 修回日期:2018-10-10 出版日期:2019-03-10 发布日期:2019-03-11
通讯作者: 李柏林
作者简介:杨飞(1994-),男,重庆垫江人,硕士研究生,主要研究方向:机器视觉、图像处理、模式识别;罗建桥(1991-),男,湖南湘潭人,博士研究生,主要研究方向:图像语义分析、机器学习;李柏林(1962-),男,广西桂林人,教授,博士生导师,博士,主要研究方向:图像处理、机器视觉。
基金资助:
四川省科技支撑计划项目（2018GZ0361）。

Railway fastener classification model based on sLDA combined with global and local constraints

YANG Fei, LUO Jianqiao, LI Bailin

School of Mechanical Engineering, Southwest Jiaotong University, Chengdu Sichuan 610031, China

Received:2018-08-24 Revised:2018-10-10 Online:2019-03-10 Published:2019-03-11
Supported by:
This work is partially supported by the Science and Technology Support Project of Sichuan Province (2018GZ0361).

摘要/Abstract

摘要： 针对监督潜在狄利克雷分布（sLDA）模型中测试图像缺乏标注，导致测试主题分布忽略目标结构的问题，提出一种结合全局和局部约束的sLDA（glc-sLDA）扣件图像分类模型。首先，人工标注训练图像，并在sLDA模型中学习得到含有结构信息的训练主题分布；然后，计算测试主题分布，将测试图像的类别概率作为全局约束，将测试图像子块与训练图像子块的主题相似程度作为局部约束；最后，以全局和局部约束的乘积为更新权值，对训练主题分布加权求和得到新的测试主题分布，并在Softmax分类器中得到测试图像的分类结果。实验结果表明，glc-sLDA模型能表达扣件结构信息，与sLDA相比，各类别的扣件图像区分性增强，分类误检率减小了55%。

关键词: 铁路扣件分类, 监督潜在狄利克雷分布, 主题模型, 单词标注, 目标结构, 更新主题分布

Abstract: Aiming at the ignorance of target structure in test topic distribution due to the lack of annotation in supervised Latent Dirichlet Allocation (sLDA) model, a sLDA fastener image classification model combined with global and local constraints (glc-LDA) was proposed. Firstly, the training images were manually labeled, and the training topic distribution with structural information was learned in sLDA model. Then, the test topic distribution was calculated to obtain the image category probabilities as global constraints, the topic similarities of training sub-blocks and test sub-blocks as local constraints. Finally, updated test topic distribution was obtained by weighted summation of training topic distribution with the product of global and local constraints as updated weights. The image category labels were obtained in Softmax classifier by the updated topics. The experimental results show that the proposed algorithm can express the structural information of fastener and compared with sLDA model, the distinction of each category of fastener images is enhanced, and the false detection rate is reduced by 55%.

Key words: railway fastener classification, supervised Latent Dirichlet Allocation (sLDA), topic model, annotation of word, target structure, update topic distribution

中图分类号:

TP391.41

杨飞, 罗建桥, 李柏林. 结合全局和局部约束的sLDA铁路扣件分类模型[J]. 计算机应用, 2019, 39(3): 888-893.

YANG Fei, LUO Jianqiao, LI Bailin. Railway fastener classification model based on sLDA combined with global and local constraints[J]. Journal of Computer Applications, 2019, 39(3): 888-893.

参考文献

[1] 王凌,张冰,陈锡爱.基于计算机视觉的钢轨扣件螺母缺失检测系统[J].计算机工程与设计,2011,32(12):4147-4150.(WANG L, ZHANG B, CHEN X A. Inspection system for loss of rail fastening nut based on computer vision[J]. Computer Engineering and Design, 2011, 32(12):4147-4150.)
[2] DOU Y, HUANG Y, LI Q, et al. A fast template matching-based algorithm for railway bolts detection[J]. International Journal of Machine Learning and Cybernetics, 2014, 5(6):835-844.
[3] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[4] 刘甲甲,李柏林,罗建桥,等.融合PHOG和MSLBP特征的铁路扣件检测算法[J].西南交通大学学报,2015,50(2):256-263.(LIU J J, LI B L, LUO J Q, et al. Railway fastener detection algorithm integrating PHOG and MSLBP features[J]. Journal of Southwest Jiaotong University, 2015, 50(2):256-263.)
[5] BLEI D M, NG A Y, JORDAN M I. Latent dirichlet allocation[J]. Journal of Machine Learning Research, 2012, 3:993-1022.
[6] 杨赛,赵春霞.基于隐含狄利克雷分配模型的图像分类算法[J].计算机工程,2012,38(14):181-183.(YANG S, ZHAO C X. Image classification algorithm based on latent Dirichlet allocation model[J]. Computer Engineering, 2012, 38(14):181-183.)
[7] 李斌,程丹,李星.基于Direct LDA的相关向量机遥感图像分类[J].信息技术,2017(4):17-20.(LI B, CHENG D, LI X. Relevant vector machine classification of hyperspectral image based on direct linear discriminant analysis[J]. Information Technology, 2017(4):17-20.)
[8] 曾康林,刘汉文.基于LDA和SVM的图像场景分类[J].中国新通信,2018,20(10):125-127.(ZENG K L, LIU H W. Image scene classification based on LDA and SVM[J]. China New Telecommunications, 2018, 20(10):125-127.)
[9] FENG H, JIANG Z, XIE F, et al. Automatic fastener classification and defect detection in vision-based railway inspection systems[J]. IEEE Transactions on Instrumentation and Measurement, 2014, 63(4):877-888.
[10] 罗建桥,刘甲甲,李柏林,等.融合纹理结构的潜在狄利克雷分布铁路扣件检测模型[J].计算机应用,2016,36(2):574-579.(LUO J Q, LIU J J, LI B L, et al. Latent dirichlet allocation model integrated with texture structure for railway fastener detection[J]. Journal of Computer Applications, 2016, 36(2):574-579.)
[11] 罗建桥,刘甲甲,李柏林,等.基于局部特征和语义信息的扣件图像检测[J].计算机应用研究,2016,33(8):2514-2518.(LUO J Q, LIU J J, LI B L, et al. Detection for railway fasteners based on local features and semantic information[J]. Application Research of Computers, 2016, 33(8):2514-2518.)
[12] 赵理君,唐娉,霍连志,等.图像场景分类中视觉词包模型方法综述[J].中国图象图形学报,2014,19(3):333-343.(ZHAO L J, TANG P, HUO L Z, et al. Review of the bag-of-visual-words models in image scene classification[J]. Journal of Image and Graphics, 2014, 19(3):333-343.)
[13] GUO Q, LI N, YANG Y, et al. Supervised LDA for image annotation[C]//Proceedings of the 2011 IEEE International Conference on Systems, Man, and Cybernetics. Piscataway, NJ:IEEE, 2011:471-476.
[14] ZHOU H. Markov weight fields for face sketch synthesis[C]//CVPR'12:Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2012:1091-1097.
[15] PENG C, GAO X, WANG N, et al. Graphical representation for heterogeneous face recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(2):301-312.
[16] WANG N, GAO X, LI J. Random sampling for fast face sketch synthesis[J]. Pattern Recognition, 2018, 76:215-227.
[17] ZHANG C, ZHU X, LI L, et al. Joint image representation and classification in random semantic spaces[J]. Neurocomputing, 2015, 156(C):79-85.
[18] OJALA T, PIETIKAINEN M, MAENPAA T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7):971-987.
[19] 李永波,李柏林,熊鹰.基于HOG特征的铁路扣件状态检测[J].传感器与微系统,2013,32(10):110-113.(LI Y B, LI B L, XIONG Y. Railway fastener state detection based on HOG feature[J]. Transducer and Microsystem Technologies, 2013, 32(10):110-113.)
[20] WANG Y, LIN X, WU L, et al. Effective multi-query expansions:collaborative deep networks based feature learning for robust landmark retrieval[J]. IEEE Transactions on Image Processing, 2017, 26(3):1393-1404.
[21] HUANG C, LUO W, XIE Y. Local-class-shared-topic latent Dirichlet allocation based scene classification[J]. Multimedia Tools and Applications, 2017, 76(14):15661-15679.

结合全局和局部约束的sLDA铁路扣件分类模型

Railway fastener classification model based on sLDA combined with global and local constraints

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	杨丰瑞, 霍娜, 张许红, 韦巍. 基于注意力机制的主题扩展情感对话生成[J]. 计算机应用, 2021, 41(4): 1078-1083.
[2]	杨威亚, 余正涛, 高盛祥, 宋燃. 基于跨语言神经主题模型的汉越新闻话题发现方法[J]. 计算机应用, 2021, 41(10): 2879-2884.
[3]	朱思淼, 魏世伟, 魏思恒, 余敦辉. 基于弹幕情感分析和主题模型的视频推荐算法[J]. 计算机应用, 2021, 41(10): 2813-2819.
[4]	尹春勇, 章荪. 面向短文本情感分类的端到端对抗变分贝叶斯方法[J]. 计算机应用, 2020, 40(9): 2536-2542.
[5]	田保军, 刘爽, 房建东. 融合主题信息和卷积神经网络的混合推荐算法[J]. 计算机应用, 2020, 40(7): 1901-1907.
[6]	徐红艳, 王丹, 王富海, 王嵘冰. 融合潜在狄利克雷分布与元路径分析的用户相关性度量方法[J]. 计算机应用, 2019, 39(11): 3288-3292.
[7]	余慧, 冯旭鹏, 刘利军, 黄青松. 聊天机器人中用户就医意图识别方法[J]. 计算机应用, 2018, 38(8): 2170-2174.
[8]	许银洁, 孙春华, 刘业政. 考虑用户特征的主题情感联合模型[J]. 计算机应用, 2018, 38(5): 1261-1266.
[9]	李琰, 刘嘉勇. 基于作者主题模型和辐射模型的用户位置预测模型[J]. 计算机应用, 2018, 38(4): 939-944.
[10]	徐立洋, 黄瑞章, 陈艳平, 钱志森, 黎万英. 基于狄利克雷多项分配模型的多源文本主题挖掘模型[J]. 计算机应用, 2018, 38(11): 3094-3099.
[11]	邓扬, 张晨曦, 李江峰. 基于弹幕情感分析的视频片段推荐模型[J]. 计算机应用, 2017, 37(4): 1065-1070.
[12]	褚征, 于炯, 王佳玉, 王跃飞. 基于LDA主题模型的移动应用相似度构建方法[J]. 计算机应用, 2017, 37(4): 1075-1082.
[13]	唐黎哲, 冯大为, 李东升, 李荣春, 刘锋. 以LDA为例的大规模分布式机器学习系统分析[J]. 计算机应用, 2017, 37(3): 628-634.
[14]	吉余岗, 李依桐, 石川. 融合异质网络与主题模型的方面分预测[J]. 计算机应用, 2017, 37(11): 3201-3206.
[15]	闫蓉, 高光来. 基于检索结果排序的伪相关反馈[J]. 计算机应用, 2016, 36(8): 2099-2102.