基于CLEAN 算法的毫米波全息二维像研究
Analysis of MMW Holographic 2D Imaging Using CLEAN Algorithm
  
DOI:
中文关键词:  毫米波, 全息成像, 二维方位像, 加窗, 二维自适应CLEAN 算法
英文关键词:millimeter wave, holographic imaging, 2D azimuth imaging, windowing, 2D adaptive CLEAN algorithm
基金项目:
作者单位
刘敏, 朱莉, 李小辉, 徐枫 南京理工大学电光学院探测与控制工程系,南京210094 
摘要点击次数: 104
全文下载次数: 
中文摘要:
      针对近程毫米波全息成像算法过程中的点函数扩散以及旁瓣过宽问题,提出了基于加窗的二维自适应CLEAN 算法。该算法提取所得“脏图冶的最亮像素点,减去该像素点以及周围小目标的部分能量,之后在清洁图像上重建目标处的精准信息,有效抑制点函数扩散与旁瓣展宽。区别于传统CLEAN 算法,二维自适应CLEAN 算法根据当前图像的能量均值实时更新减去因子,具有很强的自适应性,此外,依据图像最暗目标像素点动态获取全局噪声门限,提高了目标提取速度。通过对多个单目标进行一维与二维方位像的仿真实验,证明了算法的有效性。
英文摘要:
      To solve the problem of point spread and side lobe broadening in short range millimeter wave holographic imaging algorithm, a 2D adaptive CLEAN algorithm based on windowing is proposed. This algorithm gets the brightest pixels of the “dirty map冶, subtracting part of the energy from the pixels and the surrounding small targets, then reconstructing accurate information of the target in the clean image, effectively inhibiting the PSF and side lobe broadening. Different from the traditional CLEAN algorithm, adaptive CLEAN algorithm has very strong adaptability because it updates the minus factor in real-time according to the mean energy of the current image, in addition, improving the extraction speed on the basis of the dark image target pixels to obtain the global dynamic noise threshold. The effectiveness of the algorithm is demonstrated by simulation experiments on one-dimensional and two-dimensional azimuth images of multiple single targets.
查看全文  查看/发表评论  下载PDF阅读器
关闭