Kumar Deepak
Amity Institute of Geoinformatics & Remote Sensing (AIGIRS), Amity University, Sector 125, Gautam Buddha Nagar, Noida, Uttar Pradesh, 201303, India.
Sci Rep. 2021 Mar 18;11(1):6241. doi: 10.1038/s41598-021-85121-9.
Satellite-based remote sensing has a key role in the monitoring earth features, but due to flaws like cloud penetration capability and selective duration for remote sensing in traditional remote sensing methods, now the attention has shifted towards the use of alternative methods such as microwave or radar sensing technology. Microwave remote sensing utilizes synthetic aperture radar (SAR) technology for remote sensing and it can operate in all weather conditions. Previous researchers have reported about effects of SAR pre-processing for urban objects detection and mapping. Preparing high accuracy urban maps are critical to disaster planning and response efforts, thus result from this study can help to users on the required pre-processing steps and its effects. Owing to the induced errors (such as calibration, geometric, speckle noise) in the radar images, these images are affected by several distortions, therefore these distortions need to be processed before any applications, as it causes issues in image interpretation and these can destroy valuable information about shapes, size, pattern and tone of various desired objects. The present work aims to utilize the sentinel-1 SAR datasets for urban studies (i.e. urban object detection through simulation of filter properties). The work uses C-band SAR datasets acquired from Sentinel-1A/B sensor, and the Google Earth datasets to validate the recognized objects. It was observed that the Refined-Lee filter performed well to provide detailed information about the various urban objects. It was established that the attempted approach cannot be generalised as one suitable method for sensing or identifying accurate urban objects from the C-band SAR images. Hence some more datasets in different polarisation combinations are required to be attempted.
基于卫星的遥感在监测地球特征方面发挥着关键作用，但由于传统遥感方法存在云层穿透能力和遥感选择性持续时间等缺陷，目前人们的注意力已转向使用微波或雷达传感技术等替代方法。微波遥感利用合成孔径雷达（SAR）技术进行遥感，并且可以在所有天气条件下运行。先前的研究人员已经报道了SAR预处理对城市物体检测和制图的影响。准备高精度的城市地图对于灾害规划和应对工作至关重要，因此本研究的结果可以帮助用户了解所需的预处理步骤及其效果。由于雷达图像中存在诱导误差（如校准、几何、斑点噪声），这些图像会受到多种失真的影响，因此在进行任何应用之前都需要对这些失真进行处理，因为它会在图像解释中引发问题，并且可能会破坏有关各种所需物体的形状、大小、图案和色调的有价值信息。目前的工作旨在利用哨兵 -1 SAR数据集进行城市研究（即通过模拟滤波器特性进行城市物体检测）。该工作使用从哨兵 -1A/B传感器获取的C波段SAR数据集以及谷歌地球数据集来验证识别出的物体。据观察，改进的李滤波器在提供有关各种城市物体的详细信息方面表现良好。研究确定，所尝试的方法不能作为从C波段SAR图像中传感或识别准确城市物体的一种合适通用方法。因此，需要尝试一些不同极化组合的更多数据集。
