Current Issue Cover
  • 发布时间:
  • 摘要点击次数:
  • 全文下载次数:
  • DOI:
  • 2017 | Volume  | Number 7

()

摘 要
目的:相位解缠是InSAR干涉数据处理的关键步骤,运用统计费用网络流(SNAPHU)算法[1]解缠高分辨机载InSAR干涉数据,当数据存在由排状树木导致的相位不一致时,解缠结果出现沿相位不一致为分界的大面积相位跳变。SNAPHU为全局优化算法,在处理局部相位不一致方面仍有局限。局部相位不一致可由局部算法优化,常以残差点表述相位不一致,通过对残差点进行补偿优化局部数据。本文提出一种结合局部求优和全局求优的——结合残差点退化和统计费用网络流的机载相位解缠算法。方法:对滤波且去除平地相位后干涉图像,进行残差点退化处理。残差点退化处理包含残差点定位,及残差点补偿两部分。根据残差点及其临域像元的性质,对残差点进行补偿使其退化为非残差点,不断迭代这一过程,以减少图像中的残差点,优化局部数据。根据机载InSAR系统定标参数,修正SNAPHU算法中的参数及几何模型,使用修正后算法进行相位解缠。对解缠图像进行中值滤波,得到最终结果。结果:利用2011年四川江油地区的单轨双天线机载InSAR数据进行了试验,试验结果表明,在相位不一致,相干性低的树木繁茂区域,该算法显著缩小了解缠相位不连续区域,修正了大面积的相位跳变。结论:验证了残差点退化方法可有效解决解缠相位大面积跳变问题,且对噪声具有鲁棒性。
关键词

()

Abstract
Objective: Phase unwrapping is a key step of InSAR (Interferometric Synthetic Aperture Radar) data processing. Apply the statistical cost network flow algorithm [1] (SNAPHU) to high resolution airborne InSAR data. Once the interferogram has phase inconsistencies caused by lines of trees, the unwrapped result appears large areas of phase jump along inconsistencies correspondingly. SNAPHU is a global optimization algorithm, which has limitations in dealing with local phase inconsistencies. Local phase inconsistencies can be optimized by local algorithm. Phase inconsistencies can be described by residues. Local data can be optimized by residues compensation. This paper proposes an airborne unwrapping algorithm combines the residue degradation with the statistical-cost network-flow, which has advantages of both the local optimum and the global optimum. Method: Flatten the filtered interferogram. Then apply the residue degradation to the flattened interferogram. The residue degradation process contains residues detection and residues compensation. According to properties of residues and values of neighborhood pixels, these residues will be compensated, and keep iterating to decrease the number of residues to optimize local data. According to calibration parameters of airborne InSAR system, modify parameters and the geometric mode in statistical-cost network-flow (SNAPHU) algorithm. And the modified SNAPHU algorithm is employed for phase unwrapping. Finally, applied a median filter to the unwrapped image to get the final result. Result: The efficiency and accuracy of proposed method is tested and validated by using the single pass dual antenna airborne InSAR data covering Jiangyou, Sichuan areas in 2011. Wooded areas are low coherence and have serious phase inconsistencies, and testing results show that, the proposed algorithm significantly shrunk error unwrapped regions, and fixed large phase jump which are caused by wooded areas. Conclusion: The residue degradation process effectively solves the problem of large unwrapped phase jump, and becomes more robust to noise.
Keywords
QQ在线


关注微信