Motion Object Feature Extraction Method Based on Multi-feature Fusion

doi:10.16182/j.issn1004731x.joss.201706020

Abstract

Abstract: Motion object feature extraction is the basis of motion object classification. Traditionally motion object classification mainly depends on single feature extraction which is sensitive to the aspects like motion object detection area, angle, scale and noise disturbance, thus decreases the classification efficiency. To solve these problems and improve the robustness of the algorithms, a motion object feature extraction method based on multi-feature fusion was proposed. In this method, width height ratio feature, rotation invariant uniform local binary pattern feature and SIFT feature were considered, and by fusing them into the SVM and KNN classifier, motion object classification was carried out. Experiments prove that the motion object feature extraction method can greatly improve the average classification precision.

Key words: motion object, feature extraction, local binary pattern, scale-invariant feature transform (SIFT)

CLC Number:

TP39

Luan Xidao, Xie Yuxiang, Zhang Xin, Niu Xiao. Motion Object Feature Extraction Method Based on Multi-feature Fusion[J]. Journal of System Simulation, 2017, 29(6): 1304-1310.

References

[1] Bharath R Raj, Dhivya G.Moving object detection, classification and its parametric evaluation[C]// International Conference on Information Communication and Embedded Systems (ICICES 2014), Chennai, India. USA: Institute of Electrical and Electronics Engineers Inc., 2015: 1-6.
[2] Niluthpol Chowdhury Mithun, Nafi Ur Rashid, S M Mahbubur Rahman. Detection and classification of vehicles from video using multiple time-spatial images[J]. IEEE Transactions on Intelligent Transportation Systems (S1524-9050), 2012, 13(3): 1215-1225.
[3] Bogomolov Y, Dror G.Classification of moving targets based on motion and appearance [C]// British Machine Vision Conference, Norwich, UK. UK: BMVA Press, 2003: 429-438.
[4] 岳昊, 邵春福, 赵熠. 基于BP 神经网络的行人和自行车交通识别方法[J]. 北京交通大学学报, 2008, 32(3): 46-49.
(Yue Hao, Shao Chunfu, Zhao Yi.A study on pedestrian and cyclist recognition based on BP neural network[J]. Journal of Beijing Jiaotong University, 2008, 32(3): 46-49.)
[5] Brown L.View independent vehicle/person classification[C]// Proc. of the ACM 2nd International Workshop on Video Surveillance and Sensor Networks. USA: ACM, 2004: 114-123.
[6] 吴金勇, 赵勇, 王一科, 等. 基于多粒度感知SVM的复杂场景人车分类方法[J]. 北京大学学报(自然科学版), 2013, 49(3): 404-408.
(Wu Jinyong, Zhao Yong, Wang Yike, et al.Human and Vehicle Classification Method for Complex Scene Based on Multi-granularity Perception SVM[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2013, 49(3): 404-408.)
[7] Lun Zhang, Stan Z Li, Xiaotong Yuan, et al.Real-time Object Classiﬁcation in Video Surveillance Based on Appearance Learning[C]// IEEE Computer Vision and Pattern Recognition. USA: IEEE, 2007.
[8] Ahmed Fawzi Otoom, Hatice Gunes, Massimo Piccard.Feature extraction techniques for abandoned object classification in video surveillance[C]// IEEE International Conference of Image Processing. USA: IEEE, 2008.
[9] Jitao Sang, Zhen Lei, Shengcai Liao, et al.Adaptive object classiﬁcation in surveillance system by exploiting scene[C]// CVPR. USA: IEEE Computer Society, 2009: 1-7.
[10] Lopez, Miriam M, Marcenaro, et al. Advantages of dynamic analysis in HOG-PCA feature space for video moving object classification[C]// 2015 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2015), Brisbane, Australia. USA: IEEE, 2015: 1285-1289.
[11] 王燕玲, 李广伦, 林晓. 基于跟踪特征的多运动目标分类算法[J]. 中国体视学与图像分析, 2014, 19(2): 121-128.
(Wang Yanling, Li Guanglun, Lin Xiao.A multiple moving targets classification algorithm based on feature tracking[J]. Chinese Journal of Stereology and Image Analysis, 2014, 19(2): 121-128.)
[12] 刘素珍, 邓和林. 一种基于特征融合的运动目标跟踪算法[J]. 激光与红外, 2008, 38(8): 837-840.
(Liu Suzhen, Deng Helin.A Moving Object Tracking Algorithm Based on Feature Fusion[J]. Laser & Infrared, 2008, 38(8): 837-840.)
[13] 郑林, 韩崇昭, 左东广, 等. 基于多特征融合的运动目标识别[J]. 系统仿真学报, 2004, 16(5): 1081-1084.
(Zheng Lin, Han Chongzhao, Zuo Dongguang, et al.Moving Object Recognition Based on Multi-feature Fusion[J]. Journal of System Simulation (S1004-731X), 2004, 16(5): 1081-1084.)
[14] 纪筱鹏, 魏志强, 冯业伟, 等. 基于多特征融合的运动物体识别方法研究[J]. 哈尔滨工业大学学报, 2009, 41(3): 169-171.
(Ji Xiaopeng, Wei Zhiqiang, Feng Yewei, et al.Moving Object Recognition based on Multi-feature Fusion[J]. Journal of Harbin Institute of Technology, 2009, 41(3): 169-171.)
[15] 曹洁, 李延林. 基于信息融合的运动目标跟踪算法[J]. 电子测量与仪器学报, 2011, 25(3): 211-217.
(Cao Jie, Li Yanlin.Fusion information-based moving tracking algorithm[J]. Journal of Electronic Measurement and Instrument, 2011, 25(3): 211-217.)
[16] T Ojala, M Pietikinen.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence (S0162-8828), 2002, 24(7): 971-987.
[17] Lowe D G.Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision (S0920-5691), 2004, 60(2): 91-110.
[18] LI Fei-fei, Fergus R, Perona P.Learning generative visual models from few training examples: all incremental Bayesian approach tested on 101 object categories[C]// IEEE Conference on Computer Vision and Pattern Recognition. USA: IEEE, 2004.