基于深度特征表达与学习的视觉跟踪算法研究

李寰宇; 毕笃彦; 杨源; 查宇飞; 覃兵; 张立朝

doi:10.11999/JEIT150031

基于深度特征表达与学习的视觉跟踪算法研究

doi: 10.11999/JEIT150031

1.
(空军工程大学航空航天工程学院西安 710038)
2.
(空军工程大学空管领航学院西安 710051)

基金项目:

国家自然科学基金(61202339, 61472443)和航空科学基金(20131996013)

计量
- 文章访问数: 1353
- HTML全文浏览量: 82
- PDF下载量: 1994
- 被引次数: 0
出版历程
- 收稿日期: 2015-01-06
- 修回日期: 2015-04-28
- 刊出日期: 2015-09-19

Research on Visual Tracking Algorithm Based on Deep Feature Expression and Learning

2.
(College of Aerospace Engineering, Air Force Engineering University, Xi&rsquo

摘要

摘要: 该文针对视觉跟踪中运动目标的鲁棒性跟踪问题，将深度学习引入视觉跟踪领域，提出一种基于多层卷积滤波特征的目标跟踪算法。该算法利用分层学习得到的主成分分析(PCA)特征向量，对原始图像进行多层卷积滤波，从而提取出图像更深层次的抽象表达，然后利用巴氏距离进行特征相似度匹配估计，进而结合粒子滤波算法实现目标跟踪。结果表明，这种多层卷积滤波提取到的特征能够更好地表达目标，所提跟踪算法对光照变化、遮挡、异面旋转、摄像机抖动都具有很好的不变性，对平面内旋转也具有一定的不变性，在具有此类特点的视频序列上表现出非常好的鲁棒性。
- 视觉跟踪 /
- 深度学习 /
- 主成分分析 /
- 卷积神经网络 /
- 粒子滤波
Abstract: For the robustness of visual object tracking, a new tracking algorithm based on multi-stage convolution filtering feature is proposed by introducing deep learning into visual tracking. The algorithm uses the Principal Component Analysis (PCA) eigenvectors obtained by stratified learning, to extract the deeper abstract expression of the original image by multi-stage convolutional filtering. Then the Bhattacharyya distance is used to evaluate the similarity among features. Finally, particle filter algorithm is combined to realize target tracking. The result shows that the feature obtained by multi-stage convolution filtering can express target better, the proposed algorithm has a better inflexibility to illumination, covering, rotation, and camera shake, and it exhibits very good robustness in video sequence with such characteristics.
- Visual tracking /
- Deep learning /
- Principal Component Analysis (PCA) /
- Convolutional neural network /
- Particle filter

HTML全文

参考文献(23)

[1]	Li X, Hu W M, and Shen C H. A survey of appearance models in visual object tracking[J]. ACM Transactions on Intelligent Systems and Technology, 2013, 4(4): 5801-5848.
[2]	Hinton G E and Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[3]	Clement F, Camille C, Laurent N, et al.. Learning hierarchical features for scene labeling[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1915-1929.
[4]	Alex K, Sutskever I, and Hinton G E. ImageNet classification with deep convolutional neural networks[C]. Proceedings of Advances in Neural Information Processing Systems, Lake Tahoe, 2012: 748-764.
[5]	Zhou S S, Chen Q C, and Wang X L. Convolutional deep networks for visual data classification[J]. Neural Processing Letters, 2013, 38(11): 17-27.
[6]	Abdel-Hamid O, Mohamed A R, Jiang H, et al.. Convolutional neural networks for speech recognition[J]. ACM Transactions on Audio, Speech, and Language Processing, 2014, 22(10): 1533-1545.
[7]	Chen X Y, Xiang S M, and Li C L. Vehicle detection in satellite images by hybrid deep convolutional neural networks [J]. IEEE Transactions on Geoscience and Remote Sensing Letters, 2014, 11(10): 1797-1801.
[8]	Evgeny A S, Denis M T, and Serge N A. Comparison of regularization methods for imagenet classification with deep convolutional neural networks[J]. AASRI Procedia, 2014, 6(8): 89-94.
[9]	Baldi P and Hornik K. Neural networks and principal component analysis: learning from examples without local minima[J]. Neural Networks, 1989, 2(1): 53-58.
[10]	Ross D, Lim Jong-woo, and Lin Ruei-Sung. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(1): 125-141.
[11]	姚志均. 一种新的空间直方图相似性度量方法及其在目标跟踪中的应用[J]. 电子与信息学报, 2013, 35(7): 1644-1649.
[12]	Yao Z J. A new spatiogram similarity measure method and its application to object tracking[J]. Journal of Electronics Information Technology, 2013, 35(7): 1644-1649.
[13]	Zhang K H, Zhang L, and Yang M H. Real-time compressive tracking[C]. Proceedings of Europe Conference on Computer Vision, Florence, 2012: 864-877.
[14]	Sevilla-Lara L and Learned-Miller E. Distribution fields for tracking[C]. IEEE Conference on Computer Vision and Pattern Recognition, Colorado, 2011: 1910-1917.
[15]	Shaul O, Aharon B H, and Dan L. Locally orderless tracking[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Rhode Island, 2012: 1940-1947.
[16]	Henriques J F, Caseiro R, and Martins P. High-speed tracking with kernelized correlation filters[J]. IEEE
[17]	Transactions on Pattern Analysis and Machine Intelligence, 2015, DOI: 10.1109/TPAMI.2014.2345390.
[18]	Hare S, Saffari A, and Torr P H S. Struck：structured output tracking with kernels[C]. Proceedings of IEEE International Conference on Computer Vision, Colorado, 2011: 263-270.
[19]	Thang Ba Dinh, Nam Vo, and Medioni G. Context tracker: exploring supporters and distracters in unconstrained environments[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Colorado, 2011: 1177-1184.
[20]	Liu Bai-yang, Huang Jun-zhou, and Yang Lin. Robust tracking using local sparse appearance model and K-selection [C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Colorado, 2011: 1313-1320.
[21]	Junseok K and Kyoung M. Tracking by sampling trackers[C]. Proceedings of IEEE International Conference on Computer Vision, Colorado, 2011: 1195-1202.
[22]	Amit Adam, Ehud Rivlin, and Ilan Shimshoni. Robust fragments-based tracking using the integral histogram[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Colorado, 2006: 798-805.
[23]	Dorin Comaniciu, Visvanathan Ramesh, and Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577.

相关文章(20)

[1]	申滨, 王欣, 陈思吉, 崔太平. 基于机器学习主用户发射模式分类的蜂窝认知无线电网络频谱感知, 电子与信息学报. doi: 10.11999/JEIT191012
[2]	刘云, 薛盼盼, 李辉, 王传旭. 基于深度学习的关节点行为识别综述, 电子与信息学报. doi: 10.11999/JEIT200267
[3]	申铉京, 沈哲, 黄永平, 王玉. 基于非局部操作的深度卷积神经网络车位占用检测算法, 电子与信息学报. doi: 10.11999/JEIT190349
[4]	柯丽, 王丹妮, 杜强, 姜楚迪. 基于卷积长短时记忆网络的心律失常分类方法, 电子与信息学报. doi: 10.11999/JEIT190712
[5]	袁野, 贾克斌, 刘鹏宇. 基于深度卷积神经网络的多元医学信号多级上下文自编码器, 电子与信息学报. doi: 10.11999/JEIT190135
[6]	吕晓琪, 吴凉, 谷宇, 张明, 李菁. 基于深度卷积神经网络的低剂量CT肺部去噪, 电子与信息学报. doi: 10.11999/JEIT170769
[7]	张红颖, 王赛男, 胡文博. 基于粒子滤波与样本加权的压缩跟踪算法, 电子与信息学报. doi: 10.11999/JEIT170854
[8]	杨峰, 张婉莹. 一种多模型贝努利粒子滤波机动目标跟踪算法, 电子与信息学报. doi: 10.11999/JEIT160467
[9]	伍家松, 达臻, 魏黎明, SENHADJILotfi, 舒华忠. 基于分裂基-2/(2a)FFT算法的卷积神经网络加速性能的研究, 电子与信息学报. doi: 10.11999/JEIT160357
[10]	侯志强, 王鑫, 余旺盛, 戴铂, 金泽芬芬. 基于自适应深度稀疏网络的在线跟踪算法, 电子与信息学报. doi: 10.11999/JEIT160762
[11]	李祖贺, 樊养余, 王凤琴. YUV空间中基于稀疏自动编码器的无监督特征学习, 电子与信息学报. doi: 10.11999/JEIT150557
[12]	侯志强, 戴铂, 胡丹, 余旺盛, 陈晨, 范舜奕. 基于感知深度神经网络的视觉跟踪, 电子与信息学报. doi: 10.11999/JEIT151449
[13]	李寰宇, 毕笃彦, 查宇飞, 杨源. 一种易于初始化的类卷积神经网络视觉跟踪算法, 电子与信息学报. doi: 10.11999/JEIT150600
[14]	袁广林, 薛模根. 基于稀疏稠密结构表示与在线鲁棒字典学习的视觉跟踪, 电子与信息学报. doi: 10.11999/JEIT140507
[15]	薛模根, 朱虹, 袁广林. 基于在线判别式字典学习的鲁棒视觉跟踪, 电子与信息学报. doi: 10.11999/JEIT141325
[16]	程帅, 孙俊喜, 曹永刚, 刘广文, 韩广良. 多示例深度学习目标跟踪, 电子与信息学报. doi: 10.11999/JEIT150319
[17]	蔡自兴, 彭梦, 余伶俐. 基于时序特性的自适应增量主成分分析的视觉跟踪, 电子与信息学报. doi: 10.11999/JEIT141646
[18]	霍雷刚, 冯象初. 基于主成分分析和字典学习的高光谱遥感图像去噪方法, 电子与信息学报. doi: 10.3724/SP.J.1146.2013.01840
[19]	孙景乐, 唐林波, 赵保军, 刘晴. 基于瑞利分布的粒子滤波跟踪算法, 电子与信息学报. doi: 10.3724/SP.J.1146.2012.01225
[20]	蒋鹏, 宋华华. 基于动态分簇路由优化和分布式粒子滤波的传感器网络目标跟踪方法, 电子与信息学报. doi: 10.3724/SP.J.1146.2012.00059