Chen Shoubin, Liu Jingbin, Huang Wenchao, Chen Ruizhi
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China.
Collaborative Innovation Center of Geospatial Technology, Wuhan University, Wuhan 430079, China.
Sensors (Basel). 2018 Mar 1;18(3):739. doi: 10.3390/s18030739.
The development of Earth observation systems has changed the nature of survey and mapping products, as well as the methods for updating maps. Among optical satellite mapping methods, the multiline array stereo and agile stereo modes are the most common methods for acquiring stereo images. However, differences in temporal resolution and spatial coverage limit their application. In terms of this issue, our study takes advantage of the wide spatial coverage and high revisit frequencies of wide swath images and aims at verifying the feasibility of stereo mapping with the wide swath stereo mode and reaching a reliable stereo accuracy level using calibration. In contrast with classic stereo modes, the wide swath stereo mode is characterized by both a wide spatial coverage and high-temporal resolution and is capable of obtaining a wide range of stereo images over a short period. In this study, Gaofen-1 (GF-1) wide-field-view (WFV) images, with total imaging widths of 800 km, multispectral resolutions of 16 m and revisit periods of four days, are used for wide swath stereo mapping. To acquire a high-accuracy digital surface model (DSM), the nonlinear system distortion in the GF-1 WFV images is detected and compensated for in advance. The elevation accuracy of the wide swath stereo mode of the GF-1 WFV images can be improved from 103 m to 30 m for a DSM with proper calibration, meeting the demands for 1:250,000 scale mapping and rapid topographic map updates and showing improved efficacy for satellite imaging.
地球观测系统的发展改变了测绘产品的性质以及地图更新方法。在光学卫星测绘方法中，多线阵立体和敏捷立体模式是获取立体图像最常用的方法。然而，时间分辨率和空间覆盖范围的差异限制了它们的应用。针对这一问题，我们的研究利用宽幅图像的广泛空间覆盖和高重访频率，旨在验证宽幅立体模式进行立体测绘的可行性，并通过校准达到可靠的立体精度水平。与经典立体模式相比，宽幅立体模式具有广泛的空间覆盖和高时间分辨率的特点，能够在短时间内获取大范围的立体图像。在本研究中，高分一号（GF-1）宽视场（WFV）图像用于宽幅立体测绘，其总成像宽度为800公里，多光谱分辨率为16米，重访周期为四天。为了获取高精度数字表面模型（DSM），预先检测并补偿了GF-1 WFV图像中的非线性系统畸变。对于经过适当校准的DSM，GF-1 WFV图像宽幅立体模式的高程精度可从103米提高到30米，满足1:250,000比例尺测绘和快速地形图更新的要求，并显示出卫星成像的更高效率。
