A hybrid genetic algorithm in PBRDF modeling
文献类型：会议论文
| 作者 | Feng,Weiwei ; Wei,Qingnong ; Li,Jinhua ; Chen,Lingxin |
| 出版日期 | 2010 |
| 会议日期 | 2010-08-10 |
| 关键词 | Annealing Distribution Functions Feature Extraction Genetic Algorithms Light Scattering Parameter Estimation Reflection Refraction Simulated Annealing Surfaces |
| 页码 | 2223 - 2227 |
| 通讯作者 | Feng, W. |
| 英文摘要 | The polarized light scattered by the surface of a material contains information that can be used to describe the properties of the surface. Polarized Bidirectional Reflectance Distribution Function (PBRDF) is one of the most important factors used to represent the property of the surface. Because there is complex nonlinear relationship between the experimental results and model parameters, genetic algorithm is used to retrieve the model parameters. One drawback of the traditional genetic algorithm is that the convergence speed is slow and easy to fall into the local minimization. On the base of the traditional genetic algorithm to retrieve the parameters, simulated annealing (SA) algorithm is used to optimize the modeling of the PBRDF. The model for PBRDF and the designation of the hybrid algorithm is given in detail. For one typical painted surface, both the experiment results and the model calculation results are given. The calculation results of the model are demonstrated consistent well with the experimental results. The error convergence curve shows that, the hybrid genetic algorithm can avoid falling into the local minimization, and shorten the running time for the target function. Therefore, it is applicable used as a reference for target feature extraction and recognition in the future. © 2010 IEEE. |
| 产权排序 | (1) Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai, 264003, China; (2) Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China |
| 会议录 | Proceedings - 2010 6th International Conference on Natural Computation, ICNC 2010
 |
| 会议录出版者 | IEEE Computer Society, 445 Hoes Lane - P.O.Box 1331, Piscataway, NJ 08855-1331, United States |
| 学科主题 | 其他 |
| 语种 | 英语 |
| URL标识 | 查看原文 |
| ISSN号 | ISSN:13640321 |
| 源URL | [http://ir.yic.ac.cn/handle/133337/4796]  |
| 专题 | 烟台海岸带研究所_环境化学实验室 |
| 推荐引用方式 GB/T 7714 | Feng,Weiwei,Wei,Qingnong,Li,Jinhua,et al. A hybrid genetic algorithm in PBRDF modeling[C]. 见:. 2010-08-10. |
入库方式： OAI收割
来源：烟台海岸带研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。