基于多波段多极化仿真图像的SAR目标识别研究

发布时间：2018-03-30 17:20

  本文选题：SAR　切入点：多波段多极化　出处：《系统仿真学报》2017年10期

【摘要】：利用Creator建立目标模型,Vega的TMM工具进行纹理材质映射,基于视景仿真技术建立了多波段多极化SAR图像数据库。设计了融合遗传算法和二值粒子群的混合智能优化算法,对SAR图像的波段极化组合方式进行优化;基于未矫正和矫正后的图像分别提取Zernike矩、Gabor小波系数等构成候选特征序列,进行了多波段多极化SAR图像特征选择实验。实验结果表明,采用仿真技术建立SAR图像数据库是进行多波段多极化SAR图像识别的一种有效手段;采用优化后的特征集合能够提高多波段多极化SAR图像的识别率。
[Abstract]:Creator is used to build the target model TMM tool for texture texture mapping, and the multi-band and multi-polarization SAR image database is established based on scene simulation technology.A hybrid intelligent optimization algorithm based on genetic algorithm (GA) and binary particle swarm optimization (BPSO) is designed to optimize the band polarization combination of SAR images, and based on the uncorrected and corrected images, Zernike moment Gabor wavelet coefficients and other candidate feature sequences are extracted, respectively.The feature selection experiment of multi-band and multi-polarization SAR image is carried out.The experimental results show that the establishment of SAR image database by simulation technique is an effective method for multi-band multi-polarization SAR image recognition, and the recognition rate of multi-band and multi-polarization SAR image can be improved by using the optimized feature set.
【作者单位】： 杭州电子科技大学通信信息传输与融合技术国防重点学科实验室;杭州电子科技大学生命信息与仪器工程学院;
【基金】：国家自然科学基金(61174024,61372024) 浙江省自然科学基金(LQ13F050010)
【分类号】：TN957.52
，

本文编号：1686867