Zenithal pedestrian detection algorithm based on improved aggregate channel features and gray-level co-occurrence matrix

doi:10.11772/j.issn.1001-9081.2018051066

Abstract

Abstract: Aiming at the uniqueness of head feature and high detection error rate extracted by traditional zenithal pedestrian detection method, a multi-feature fusion zenithal pedestrian detection algorithm based on improved Aggregate Channel Feature (ACF) and Gray-Level Co-occurrence Matrix (GLCM) was proposed. Firstly, the extracted Hue, Sturation, Value (HSV) color features, gradient magnitude and improved Histogram of Oriented Gradients (HOG) feature were combined into ACF descriptor. Then, the improved GLCM parameter descriptor was calculated by the window method, and the texture features were extracted. The co-occurrence matrix feature descriptor was obtained by concatenating the feature vectors of each window. Finally, the aggregate channel and co-occurrence matrix features were input into Adaboost for training to get the classifier, and the final results were obtained by detection. The experimental results show that, the proposed algorithm can effectively detect targets in the presence of interference background, and improve the detection precision and recall.

Key words: zenithal pedestrian detection, Aggregate Channel Feature (ACF), Gray-Level Co-occurrence Matrix (GLCM), Histogram of Oriented Gradients (HOG), Adaboost

摘要： 针对传统俯视行人检测方法提取的头部特征单一、检测错误率高的问题，提出了结合改进聚合通道特征（ACF）和灰度共生矩阵（GLCM）的俯视行人检测算法。首先，将提取到的HSV颜色特征、梯度幅值大小以及改进后的梯度方向直方图（HOG）特征组合成ACF描述子；然后，采用窗口法计算改进的GLCM参数描述子，提取纹理特征，串联每个窗口的特征向量得到共生矩阵特征描述子；最后，将聚合通道和共生矩阵特征分别输入Adaboost训练得到分类器，并进行检测得到最终结果。实验结果表明，所提算法能在干扰背景存在的情况下有效检测目标，提高了检测的准确率和召回率。

关键词: 俯视行人检测, 聚合通道特征, 灰度共生矩阵, 梯度方向直方图, Adaboost

CLC Number:

TP391.4

LI Lin, ZHANG Tao. Zenithal pedestrian detection algorithm based on improved aggregate channel features and gray-level co-occurrence matrix[J]. Journal of Computer Applications, 2018, 38(12): 3367-3371.

李琳, 张涛. 结合改进聚合通道特征和灰度共生矩阵的俯视行人检测算法[J]. 计算机应用, 2018, 38(12): 3367-3371.

References

[1] QIAN X, YU X, FA C. The passenger flow counting research of subway video based on image processing[C]//Proceedings of the 29th Chinese Control and Decision Conference. Piscataway, NJ:IEEE, 2017:5195-5198.
[2] AGGARWAL J K, RYOO M S. Human activity analysis:a review[J]. ACM Computing Surveys, 2011, 43(3):Article No. 16.
[3] CHEN T H, CHEN T Y, CHEN Z X. An intelligent people-flow counting method for passing through a gate[C]//Proceedings of the 2006 IEEE Conference on Robotics, Automation and Mechatronics. Piscataway, NJ:IEEE, 2006:1-6.
[4] YU H B, HE Z W, LIU J L. A vision-based method to estimate passenger flow in bus[C]//Proceedings of the 2007 International Symposium on Intelligent Signal Processing and Communication Systems. Piscataway, NJ:IEEE, 2007:654-657.
[5] LI J, HE Q, YANG L, et al. Pedestrian detection and counting in crowded scenes[C]//Proceedings of the 2016 International Conference on Green Intelligent Transportation System and Safety, LNEE 419. Berlin:Springer, 2016:495-511.
[6] 牛胜石,毛晓晖,侯建华,等.基于Adaboost和SVM的人头实时检测[J].微型机与应用,2010(13):33-36.(NIU S S, MAO X H, HOU J H, et al. Realtime head detection based on Adaboost and SVM[J]. Microcomputer & Its Applications, 2010(13):33-36.)
[7] 汤石晨,陈锻生.基于光流法和行人面积特征的人数统计方法研究[J].图学学报,2013,34(1):139-144.(TANG S C, CHEN D S. A method for people counting based on optical flow and pedestrian size[J]. Journal of Graphics, 2013, 34(1):139-144.)
[8] LI B, ZHANG J, ZHANG Z, et al. A people counting method based on head detection and tracking[C]//Proceedings of the 2014 International Conference on Smart Computing. Piscataway, NJ:IEEE, 2014:136-141.
[9] TANG C H, CHEN Q J. Zenithal people counting using histogram of oriented gradients[C]//Proceedings of the 5th International Congress on Image and Signal Processing. Piscataway, NJ. IEEE, 2012:946-951.
[10] 刘恋,贺赛先,熊杰.一种基于多特征融合的头部检测算法[J].电光与控制,2016,23(9):106-110.(LIU L, HE S X, XIONG J. A head detection algorithm based on multi-feature fusion[J]. Electronics Optics & Control, 2016, 23(9):106-110.)
[11] ZENG C B, MA H D. Robust Head-shoulder detection by PCA-based multilevel HOG-LBP detector for people counting[C]//Proceedings of the 20th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2010:2069-2072.
[12] DOLLÁR P, APPEL R, BELONGIE S, et al. Fast feature pyramids for object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(8):1532-1545.
[13] HARALICK R M, SHANMUGAM K, DINSTEI I. Texture features for image classification[J]. IEEE Transactions on Systems, Man & Cybernetics, 1973, SMC-3(6):610-621.
[14] SMITH A R. Color gamut transform pairs[C]//Proceedings of the 5th Conference on Computer Graphics and Interactive Techniques. New York:ACM, 1978:12-19.
[15] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2005:886-893.
[16] BOURDEV L, BRANDT J. Robust object detection via soft cascade[C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2005:236-243.
[17] YANG B, YAN J J, LEI Z, et al. Aggregate channel features for multi-view face detection[C]//Proceedings of the 2014 IEEE International Joint Conference on Biometrics. Piscataway, NJ:IEEE, 2014:1-8.
[18] OJALA T, PIETIKAINEN M, HARWOOD D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions[C]//Proceedings of the 12th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 1994:582-585.
[19] VIOLA P, JONES M J. Robust real-time object detection[J]. International Journal of Computer Vision, 2004, 57(2):137-154.
[20] FELZENSZWALB P F, GIRSHICK R B, MCALLESTER D, et al. Object detection with discriminatively trained part-based models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9):1627-1645.
[21] GAN G L, CHENG J. Pedestrian detection based on HOG-LBP feature[C]//Proceedings of the 7th International Conference on Computational Intelligence and Security. Piscataway, NJ:IEEE, 2011:1184-1187.
[22] 唐春晖.一种基于多特征融合的单目俯视行人检测[J].系统仿真学报,2016,28(9):2146-2153.(TANG C H. Zenithal pedestrian detection using multiple feature fusion in monocular vision[J]. Journal of System Simulation, 2016, 28(9):2146-2153.)
[23] 王强,冯燕.基于颜色和形状信息的快速人数统计方法[J].计算机测量与控制,2010, 18(9):2157-2159.(WANG Q, FENG Y. A fast people counting algorithm based on fusion of color and shape information[J]. Computer Measurement & Control, 2010, 18(9):2157-2159.)
[24] DOLLÁR P, TU Z W, PERONA P, et al. Integral channel features[C]//Proceedings of the 2009 British Machine Vision Conference. Durham, UK:BMVA Press, 2009:91.1-91.11.
[25] 栾书鹏.基于图块和二阶统计特征的行人检测[J].电子设计工程,2016,24(20):190-193.(LUAN S P. Pedestrian detection based on blocks-graphs and second-order statistics[J]. Electronic Design Engineering, 2016, 24(20):190-193.)