基于分等级对象语义图模型的复杂目标自动识别方法研究

孙显; 付琨; 王宏琦

doi:10.3724/SP.J.1146.2010.00965

基于分等级对象语义图模型的复杂目标自动识别方法研究

doi: 10.3724/SP.J.1146.2010.00965

孙显^{* 付琨王宏琦,},
付琨,
王宏琦

基金项目:

国家自然科学基金(41001285)和国家863计划项目(2006AA12Z149)资助课题

计量
- 文章访问数: 3216
- HTML全文浏览量: 135
- PDF下载量: 778
- 被引次数: 0
出版历程
- 收稿日期: 2010-09-07
- 修回日期: 2010-12-17
- 刊出日期: 2011-05-19

Hierarchical Objects Semantic Graph Based Hybrid Learning Method for Automatic Complicated Objects Recognition

Sun Xian^{* 付琨王宏琦
,},
Fu Kun,
Wang Hong-Qi

摘要

摘要: 目标自动识别是图像处理领域的研究热点。针对现有方法的不足，该文提出一种新的基于分等级对象语义图模型的复杂目标自动识别方法。该方法通过构建分等级对象语义图模型增强对目标与背景间、目标部件间语义约束的利用，引入置信对象网络统计局部特性，利用消息机制传递对象间相互影响，实现概率语义分析。训练中还将产生式和判别式方法结合，提高了目标识别的准确度。在自然和遥感部分目标类别数据集上的测试结果表明，该方法能完成对多种类型和复杂结构目标的识别和提取，具有一定的实用价值。
- 图像处理 /
- 目标识别 /
- 分等级语义 /
- 混合学习 /
- 置信消息
Abstract: Automatic objects recognition is a key issue in image processing area. A new hierarchical objects semantic graph based hybrid learning method is proposed to recognize targets in complicated images. This method builds a hierarchical semantic graph model to reinforce the semantic constraints among targets, background, and components in images. And it also proposes a belief objects network to improve the utilization of spatial information, by using local classifier to calculate objects properties and using belief messages to propagate the objects relationships. Besides, the method uses discriminative learning and generative learning interleavely to improve the training error, memory usage and recognition efficiency. Experimental results demonstrate that the proposed method is meaningful and helpful for image understanding.
- Image processing /
- Object recognition /
- Hierarchical semantic /
- Hybrid learning /
- Belief message

HTML全文

参考文献(1)

Weber M, Welling M, and Perona P. Towards automatic discovery of object categories[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, USA, 2000: 2101-2108.[2] 张正, 王宏琦, 孙显, 巩大亮, 胡岩峰. 基于部件的自动目标检测方法研究[J]. 电子与信息学报, 2010, 32(5): 1017-1022.Zhang Zheng, Wang Hong-qi, Sun xian, Gong Da-liang, and Hu Yan-feng, et al.An automatic method for targets detection using a component-based model[J]. Journal of Electronics Information Technology, 2010, 32(5): 1017-1022.[3] Sivic J, Russel B, Efros A, Zisserman A, and Freeman W. Discovering object categories in image collection[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2005: 370-377.[4] Opelt A, Fussenegger M, Pinz A, and Auer P. Weak hypotheses and boosting for generic object detection and recognition[C]. Proceedings of the 8th European Conference on Computer Vision, Czech Republic, 2004, Vol. 2: 71-84.[5] Leibe B A and Leonardis B Schiele. Combined object categorization and segmentation with an implicit shape model[C]. In ECCV'04 Workshop on Statistical Learning in Computer Vision, Czech Republic, 2004: 17-32.[6] Fergus R, Perona P, and Zisserman A. Object class recognition by unsupervised scale-invariant learning[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Madison, Wisconsin, USA, 2003, 2: 264-271.[7] Vijayanarasimhan S and Grauman K. Keywords to visual categories: multiple-instance learning for weakly supervised object categorization[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, USA, 2008: 1-8. [8] Kokkinos I and Yuille A. HOP: hierarchical object parsing[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, USA, 2009: 802-809.[9] Sun Xian, Wang Hong-qi, and Fu Kun. Automatic detection of geospatial objects using taxonomic semantics[J]. Geoscience and Remote Sensing Letters, 2010, 7(1): 23-27.[10] Glocker B, Zikic D, Komodakis N, and Paragios N. Linear image registration through MRF optimization[C]. IEEE International Symposium on Biomedical Imaging: from Nano to Macro, USA, 2009: 422-425.[11] Pal C, Sutton C, and McCallum A. Sparse forward-backward using minimum divergence beams for fast training of conditional random fields[C]. Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, Toulouse, 2006: 14-19.[12] Kai N, Kannan A, Criminisi A, and Winn J. Epitomic location recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12): 2158-2167.[13] Rong X, Wujun L, Yuandong T, and Xiaoou T. Joint boosting feature selection for robust face recognition[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, USA, 2006: 1415-1422.[14] Chi-Yun Y, Burn K, and Wermter S. A neural wake-sleep learning architecture for associating robotic facial emotions[C]. IEEE World Congress on Computational Intelligence Neural Networks, Hong Kong, 2008: 2715-2721.

施引文献

资源附件(0)

访问统计