全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种基于图像区域系综分类的室外场景理解方法

DOI: 10.11834/jig.2004012271

Keywords: 场景理解,多层感知机,基于图像,分类器,场景图,视觉特征,图像分割,系综,外场,对应关系

Full-Text   Cite this paper   Add to My Lib

Abstract:

多层感知机分类器是一种有效的数据分类方法,但其分类性能受训练样本空间的限制。通过多层感知机分类器系综提高室外场景理解中图像区域的分类性能,提出了一种自动识别室外场景图像中多种景物所属概念类别的方法。该方法首先提取图像分割区域的低层视觉特征,然后基于系综分类方法建立区域视觉特征和语义类别的对应关系,通过合并相同标注区域,确定图像中景物的高层语义。对包含5种景物的150幅图像进行测试,识别率达到了87%。与基于多层感知机方法的实验结果相比,本文提出的方法取得了更好的性能,这表明该方法适合于图像区域分类。此外,系综方法还可以推广到其他的分类问题。

References

[1]  Shapiro Linda, Stockman George. Computer Vision [ M ].Englewood Cliffs, NJ, Prentice Hall, 2000: 13~ 20.
[2]  Marti Joan. A new approach to outdoor scene description based on learning and top-down segmentation [J]. Image and Vision Computing, 2001, 19(14): 1041~1055.
[3]  Duds R, Hart P. Pattern Classification(2nd Edition)[M]. New York: John Wiley & Sons, 2001: 475~478.
[4]  Thomas G, Dietterich. Machine learning research: four current directions [J ]. Artificial Intelligence Magazine, 1997, 15 (4):97~136.
[5]  Lucchese L, Mitrs S K. Advances in color image segmentation [A]. In: Global Telecommunications Conference [C]. Rio de Janeiro, Brazil, 1999: 2038~2044.
[6]  Comaniciu Dorin, Meer Peter. Robust analysis of feature spaces; color image segmentation[A]. In: IEEE Conference on Computer Vision and Pattern Recognition[C]. San Juan, Puerto Rico, 1997: 750~757.
[7]  Hu M K. Visual pattern recognition by moment invariant [J].IEEE Transactions on Information Theory, 1962, 8 (2): 179 ~187.
[8]  Kodratoff Y, Moscatelli S. Machine learning for object recognition and scene analysis [ J]. International Journal of Pattern Recognition and Artificial Intelligence, 1994, 8 (1):259~304.
[9]  Song Y, Zhang A. Analyzing scenery images by monotonic tree[J]. ACM Multimedia Systems, 2003, 8(6): 495~511.
[10]  Campbell N W, Mackeown W P. Interpreting image databases by region classification[J]. Pattern Recognition, 1997, 30(4):555~563.
[11]  Hiroki Hayashi, Mineichi Kudo. Fast labelling of natural scenes using enhanced knowledge[J]. Pattern Analysis & Applications,2001, 4(1): 20~27.
[12]  Thomas G, Dietterich. Ensemble methods in machine learning[A]. In: First International Workshop on Multiple Classifier Systems [C], Cagliari, Italy, 2000:1~ 15.
[13]  Luo Jiebo, Guo Chengen. Perceptual grouping of segmented regions in color images[J]. Pattern Recognition, 2003, 36(12):2781~2792.
[14]  Comaniciu Dorin. Mean shift: a robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603~619.
[15]  Manjunath B S, Ma W Y. Texture features for browsing and retrieval of image data [J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 1996, 18(8): 837~842.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133