基于墙体补偿与多视角融合的穿墙雷达建筑物布局成像算法

Building Layout Imaging Algorithm Based on Wall Compensation and Multi-view Fusion in Through-Wall Radar

  • 摘要: 穿墙雷达能够对建筑进行布局成像,并准确获取建筑内部结构信息,已广泛应用于灾害搜救、反恐作战等领域。针对建筑物墙体穿透导致的墙体展宽、位置偏移和墙体多径杂波导致的多径鬼影等成像问题,本文提出了一种基于墙体补偿与多视角融合的建筑布局成像算法。首先,对雷达回波的后向投影成像结果应用Radon变换,通过恒虚警率检测提取峰值,提取墙体位置信息,以墙体前后表面中心为分界线将图像分割为两个子图像;其次,对包含偏移墙体的子图像进行自适应补偿,矫正墙体位置;最后,对多个视角成像结果进行图像融合,抑制多径鬼影。仿真结果表明,该方法可以有效解决图像中墙体展宽、位置偏移和多径鬼影问题,得到正确清晰的建筑布局图像。

     

    Abstract: Through-wall radar can image the building layout and obtain the accurate internal structure information, which has been widely used in disaster search and rescue, anti-terrorism operations, etc. Aiming at the imaging problems of the wall spreading, wall position shift caused by the penetration of the building wall and the multipath ghost caused by the wall multipath clutter, a building layout imaging algorithm based on wall compensation and multi-view fusion is proposed in this paper. Firstly, the Radon transformation is applied to the result of back projection image, the peak is extracted through constant false alarm rate detection to get the wall position information, and the image is split into two sub-images with the center of the front and back surface of the wall; secondly, the position of the wall is compensated adaptively with the sub-image containing the offset; finally, the images from multiple perspectives are fused to suppress multipath ghosts. The simulation results show that this method can effectively correct the wall position, suppress multipath ghosts and obtain a correct and clear building layout image.

     

/

返回文章
返回