论文部分内容阅读
目的 Landsat-8传感器采用线阵推扫式成像,通过分景处理获得标准景产品。面对大范围遥感数据需求,长条带数据具有更高的应用价值。长条带数据的一般获取方法是通过拼接标准景正射产品的方式,该方法处理效率不高,并且当条带中某景由于云覆盖量较多无法生产正射产品时,将无法通过多景拼接的方式生成相应的长条带正射产品。针对此方法存在的问题,本文提出一种Landsat-8长条带影像的正射校正方法。方法以长条带影像为整体,通过控制点匹配、长条带几何精校正和正射校正3个部分,直接生成长条带级别的正射校正产品。在长条带几何精校正过程中,提出了Landsat-8的基于轨道约束的长条带平差模型及控制点优化选取方法。结果在15景的条带长度范围内,本文方法生成的长条带正射产品的几何精度在12 m以内,满足Landsat-8正射产品的精度要求,且处理效率相较于标准景拼接方式提升1倍左右;对于云覆盖较多的景,利用连续3景进行长条带正射校正,可以得到与标准景正射产品几何精度相当的长条带正射产品。结论本文提出的长条带处理方法,在一定长度范围内,能够得到满足Landsat-8正射产品的精度要求的长条带正射产品,处理效率得到较大提高,且能够有效克服传统方法中云覆盖情形对长条带产品获取的限制。
Purpose The Landsat-8 sensor uses a linear array of push-broom imaging to obtain a standard landscape via scene-split processing. The face of large-scale remote sensing data needs, long strip data has a higher value. The general method of obtaining long stripe data is to stitch the standard ortho shot products by stitching, which is not efficient to process and can not pass more when a certain scene in the stripe can not produce ortho products because of a large amount of cloud coverage King splicing way to generate the corresponding long strip orthophoto products. Aiming at the problem of this method, this paper presents a method of orthorectification of Landsat-8 long strip images. The method takes the long strip image as a whole and directly generates long strip level orthorectification products through control point matching, long strip geometry fine correction and orthorectification. In the process of geometric correction of long strip geometry, Landsat-8’s strip-based strip adjustment model and optimal control point selection method are proposed. Results The geometric accuracy of long-strip orthophoto products generated by this method is within 12 m within the range of the stripe length of 15 scenes, which meets the accuracy requirements of Landsat-8 ortho-products and the processing efficiency is better than that of the standard splicing scheme Increase about 1 times; for the cloud cover more King, the use of continuous 3 scene for long shot orthorectification can be obtained with the standard Jing Zheng products geometrical accuracy of long strip orthophotonics. Conclusion The long strip processing method proposed in this paper can obtain the long strip orthophotonical products that meet the accuracy requirements of Landsat-8 ortho-products over a certain length, which can greatly improve the processing efficiency and can effectively overcome the disadvantages of traditional methods Limitations on long strip products obtained by cloud coverage.