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...

基于分水岭变换和遗传算法的自动目标检测

DOI: 10.11834/jig.20080915

Keywords: 目标检测,分水岭变换,区域邻接图,能量函数优化,遗传算法

Full-Text   Cite this paper   Add to My Lib

Abstract:

提出了一种新的直接从图像中检测特定目标的算法。算法分为3个步骤。首先用分水岭变换对图像作“过分割”,得到标注的积水盆地和堤坝。实际目标可能是其中一个或多个盆地的组合。然后对盆地进行初步筛选,删除不可能是目标的盆地,并用区域邻接图(RAG)表示剩下的盆地。最后针对问题特点定义能量函数,提出一种启发式遗传算法,用来在RAG中检测使得能量函数极小化的子图,子图对应的区域就是目标。实验结果表明了该算法的有效性。

References

[1]  王润生,图像理解,长沙:国防科技大学出版社,1995.
[2]  Tu Z,Chen X,Yuille A,Image parsing:segmentation,detection,and object recognition,Nice,France,2003.
[3]  Chen J H,Chen C S,Chen Y S,Fast algorithm for robust template matching with M-estimators,IEEE Transactions on Signal Processing,2003(1).
[4]  Lai K F,Chin R T,Deformable contours:modeling and extraction,IEEE Transactions on Pattern Analysis and Machine Intelligence,1995(11).
[5]  Bravo M J,Farid H,Object segmentation by top-down processes,Visual Cognition,2003(4).
[6]  Lee K M,Street W N,Generalized Hough transforms with flexible templates,Las Vegas,Nevada,USA,2000.
[7]  Roerdink JBTM,Meijster A,The watershed transform:definitions,algorithms and parallelization strategies,Fundamental Information,2000(1).
[8]  Koza J R,Survey of genetic algorithms and genetic programming,San Francisco,California,USA,1995.
[9]  Llado J,Marti E,Jose J,Symbol recognition by subgraph matching between region adjacency graphs,IEEE Transactions on Pattern Analysis and Machine Intelligence,2001(10).
[10]  Vincent L,Soille P,Watersheds in digital spaces:an efficient algorithm based on immersion simulations,IEEE Transactions on Pattern Analysis and Machine Intelligence,1991(6).
[11]  Zhang M,Ciesielski V,Neural networks and genetic algorithms for domain independent muhiclass object detection,International Journal of Computational Intelligence and Applications,2004(1).

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133