基于主动式边界基元模型的多类目标识别方法

doi:10.7523/j.issn.2095-6134.2009.4.011

中国科学院大学学报 ›› 2009, Vol. 26 ›› Issue (4): 503-512.DOI: 10.7523/j.issn.2095-6134.2009.4.011

基于主动式边界基元模型的多类目标识别方法

孙显^1,2, 胡岩峰¹, 王宏琦¹

1. 中国科学院电子学研究所，北京 100190;
2. 中国科学院研究生院，北京 100049

收稿日期:2008-12-02 修回日期:2009-04-10 发布日期:2009-07-15
通讯作者: 孙显
基金资助:
国家自然科学基金(40871209)、国家863计划(2006AA12Z149)和中国科学院电子学研究所青年创新基金资助 

Multi-categorical object recognition using method based on active contour basis model

SUN Xian^1,2, HU Yan-Feng¹, WANG Hong-Qi¹

1. Institute of Electronic, Chinese Academy of Sciences, Beijing 100190, China;
2. Graduate University of the Chinese Academy of Sciences, Beijing 100049, China

Received:2008-12-02 Revised:2009-04-10 Published:2009-07-15

摘要/Abstract

摘要：

针对形状特征,提出了一种基于主动式边界基元模型的多类目标自动识别方法. 该方法以主动式边界基元为基础构建字典,可准确描述各类目标的形状结构, 不受尺度、旋转等变化的影响;然后,综合分析上下文信息进行概率学习,采用级联框架和Bootstrap动态采样训练最优边界分类器,实现目标的类别识别和位置定位,并可获取精确形状. 实验结果表明,该方法能有效提取多种类型和复杂结构的目标,具有较强的实用价值.

关键词: 目标识别, 边界基元, 概率学习, 形状特征

Abstract:

A new multi-categorical object recognition method based on the active contour basis model is proposed. The method builds a class-specific codebook of active contour bases, which is robust to scale variation and pose changes. Probabilistic learning by analyzing contextual information is performed using cascaded frame and boot strap dynamic sampling. A classifier is trained to determine the object categories and exact regions. Experimental results demonstrate that the proposed method achieves high efficiency in extracting manifold and complicated objects.

Key words: object recognition, contour basis, probabilistic learning, shape feature

中图分类号:

TP391

孙显, 胡岩峰, 王宏琦. 基于主动式边界基元模型的多类目标识别方法[J]. 中国科学院大学学报, 2009, 26(4): 503-512.

SUN Xian, HU Yan-Feng, WANG Hong-Qi. Multi-categorical object recognition using method based on active contour basis model[J]. , 2009, 26(4): 503-512.

参考文献

[1] Leibe B, Leonardis A, Schiele B. Robust object detection with interleaved categorization and segmentation
[J]. International Journal of Computer Vision Special Issue on Learning for Recognition and Recognition for Learning, 2008, 77(1): 259-289.

[2] Felzenszwalb P F, Schwartz J D. Hierarchical matching of deformable shapes //Proc IEEE Conf Computer Vision and Pattern Recognition. 2007: 1-8.

[3] Zhang X Q, Guo M M, Tang Y. A new geometric feature shape descriptor
[J]. Computer Engineering and Applications, 2007, 43(29):90-92.

[4] Kumar M P, Torr P H S, Zisserman A. Extending pictorial structures for object recognition //Proc British Machine Vision Conference. 2004: 1-8.

[5] Fergus R, Perona P, Zisserman A. Weakly supervised scale-invariant learning of models for visual recognition
[J]. International Journal of Computer Vision, 2007, 71(3): 273-303.

[6] Opelt A, Pinz A, Zisserman A. Incremental learning of object detectors using a visual shape alphabet //Proc IEEE Conf Computer Vision and Pattern Recognition. 2006: 3-10.

[7] Shotton J, Blake A, Cipolla R. Multi-scale categorical object recognition using contour fragments
[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2008, 30(7):1270-1281.

[8] Borgefors G. Hierarchical chamfer matching: A parametric edge matching algorithm
[J]. IEEE PAMI, 1988, 10(6):849-865.

[9] Scassellati B, Alexopoulos S, Flickner M. Retrieving images by 2d shape: A comparison of computation methods with human perceptual judgments //Storage and Retrieval for Image and Video Databases. 1994: 2-14.

[10] Yin K, He J Z. Application of ISODATA iterative algorithm based on image search of the content
[J]. Journal of Anhui University of Technology, 2006, 23(4):440-443(in Chinese). 尹柯, 何建忠. ISODATA高效迭代算法在基于内容的图像检索中的应用
[J]. 安徽工业大学学报, 2006, 23(4):440-443.

[11] Kumar M, Torr P, Zisserman A. Learning layered motion segmentations of video //IEEE International Conf on Computer Vision. 2007: 33-40.

[12] Billings S A, Wei H L. Sparse model identification using a forward orthogonal regression algorithm aided by mutual information
[J]. IEEE Trans on Neural Networks, 2007, 18(1):306-310.

[13] Torralba A, Murphy K P, Freeman W T. Sharing visual features for multiclass and multiview object detection
[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2007, 19(5):854-869.

[14] Tian C N, Gao X B, Li J. An example selection method for active learning based on embedded bootstrap algorithm
[J]. Journal of Computer Research and Development, 2006,43(10):1706-1712(in Chinese). 田春娜, 高新波, 李洁. 基于嵌入式Bootstrap的主动学习示例选择方法
[J]. 汁算机研究与发展, 2006, 43(10):1706-1712.

[15] Rother C, Kolmogorov V, Blake A. GrabCut—interactive foreground extraction using iterated graph cuts
[J]. ACM Trans Graphics, Los Angeles, 2004, 23(3): 309-314.

[16] http://www.pascal-network.org/challenges/VOC/

[17] Fawcett T. An introduction to ROC analysis //Pattern Recognition Letter. 2006: 861-874.

基于主动式边界基元模型的多类目标识别方法

Multi-categorical object recognition using method based on active contour basis model

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	王泽隆, 徐向辉, 张雷. 基于仿真SAR图像深度迁移学习的自动目标识别[J]. 中国科学院大学学报, 2020, 37(4): 516-524.
[2]	邹浩, 林赟, 洪文. 基于多角度合成SAR图像的目标识别性能分析[J]. 中国科学院大学学报, 2019, 36(2): 226-234.
[3]	马娇, 董勇伟, 李原, 李凌霄, 杨杰芳. 多旋翼无人机微多普勒特性分析与特征提取[J]. 中国科学院大学学报, 2019, 36(2): 235-243.
[4]	李松, 魏中浩, 张冰尘, 洪文. 深度卷积神经网络在迁移学习模式下的SAR目标识别[J]. 中国科学院大学学报, 2018, 35(1): 75-83.
[5]	肖垚, 刘畅. 基于稀疏求解的改进PCA方法在SAR目标识别中的应用[J]. 中国科学院大学学报, 2018, 35(1): 84-88.
[6]	向卫力, 李晓辉, 周勇胜, 李传荣, 唐伶俐. 一种鲁棒的多尺度稀疏表示SAR目标识别方法[J]. 中国科学院大学学报, 2017, 34(1): 99-105.