首页> 外文会议>International Conference on Environmental and Engineering Geophysics >Research about a method of enhancing SAR (Synthetic Aperture Radar) image spatial resolution-based on the theory of interleaved sampling CCD image reconstruction
【24h】

Research about a method of enhancing SAR (Synthetic Aperture Radar) image spatial resolution-based on the theory of interleaved sampling CCD image reconstruction

机译:基于交错采样CCD图像重建理论的SAR图像空间分辨率增强方法研究

获取原文

摘要

SAR (Synthetic Aperture Radar) image has characteristics of all weather,all-day,intensive penetrability etc.,which can play an important role in many research fields.However,the recognition capability of spatial information is restricted because of large quantity noise in SAR image.Based on the theory of interleaved sampling CCD image reconstruction,we processed four multi-temporal ERS LR (Low Resolution)SAR images of same place to reduce noise and enhance resolution in this paper,and then acquired a HR(High Resolution) image.The method used indicates the spatial resolution can be enhanced by use of time frames.The process changing time resolution into spatial resolution comes true.The obtaining course of SPOT5 2.5m panchromatic band makes full use of the theory above.The paper applied the theory into the anaphase process of seasoned image,and didn't modify the imaging process.We acquire the HR image by man-induced resampling and reconstructing image.The detailed process includes 3 main parts: image preprocessing,interleaved sampling and interpolating,and resolution improving.In the image preprocessing part,the four multi-temporal LR SAR images are intercorrected to high-accuracy,and the histogram stretching of them is completed to reduce the contrast of DN;in the interleaved sampling and interpolating part,the average DN value of four images are imparted into the overlap region,which preserves the information of four images and can reduce noise partly;in the resolution improving part,many calculations are applied to the images which interpolated and reconstructed.We choose a better calculation to decrease fuzziness and reduce noises and enhance resolution.Compared to the original LR image,the HR image processed is clearer and some ground targets are identified easily in it where noises are smoothed.Then,the LR image,intermediate image,and HR image are fused with TM image and the fusion images are analyzed.Both the single HR images and the fusion images prove that the method is preferable and feasible.Besides the SAR images,it can be used in many other RS images,which makes RS images widespread availability.The way can be popularized and enhanced for further.
机译:合成孔径雷达(SAR)图像具有全天候,全天,高穿透性等特点,在许多研究领域中都起着重要作用。但是,由于SAR中存在大量噪声,使得空间信息的识别能力受到限制。基于交错采样CCD图像重建的理论,本文对同一位置的四幅多时相ERS LR(低分辨率)SAR图像进行处理,以减少噪声并提高分辨率,然后获得HR(高分辨率)图像。该方法表明可以通过使用时间帧来提高空间分辨率。将时间分辨率转换为空间分辨率的过程得以实现。SPOT52.5m全色波段的获取过程充分利用了上述理论。进入经验丰富的图像的后期过程,并且不改变成像过程。我们通过人为的重采样和重建图像来获取HR图像。详细过程包括3个步骤。 n个部分:图像预处理,交错采样和插值以及分辨率提高。在图像预处理部分中,对四个多时间LR SAR图像进行了相互校正以达到高精度,并且完成了它们的直方图拉伸以减小DN的对比度。 ;在交错采样和内插部分中,将四个图像的平均DN值赋予重叠区域,从而保留了四个图像的信息并可以部分减少噪点;在分辨率提高部分,对该图像进行了许多计算进行插值和重建。我们选择了更好的计算方法,以减少模糊性并减少噪声并提高分辨率。与原始LR图像相比,处理后的HR图像更清晰,并且在其中可以平滑噪声的位置轻松识别出一些地面目标。图像,中间图像和HR图像与TM图像融合,并对融合图像进行分析。单个HR图像和融合图像均证明了除SAR图像外,它还可以用于许多其他的RS图像中,从而使RS图像具有广泛的可用性。该方法可以进一步推广和增强。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号