首页> 外文期刊>IEEE Geoscience and Remote Sensing Letters >An Algorithm to Extract More Accurate Slopes From DEMs
【24h】

An Algorithm to Extract More Accurate Slopes From DEMs

机译:一种从DEM提取更准确坡度的算法

获取原文
获取原文并翻译 | 示例
           

摘要

In this letter, a new algorithm Quad is proposed to improve slope accuracy. According to the traditional Horn algorithm, the importance of neighboring cells depends on their distance from the center, but strangely, no consideration is given to the center cell itself. The loss of self-elevation leads to unpredictable distortions, which can cause errors. To make up for this kind of shortfall, the Quad algorithm uses the 2 $times$ 2 cell neighborhood model instead of the 3 $times$ 3 one so that the center cell's own elevation becomes the most important factor during calculation. While the horizontal and vertical cells are twice as important as the corn cell, the center cell gets twice larger than the corn one. The comparison between Horn and Quad is based on three different artificial surfaces, in which we can obtain the true slope value by derivation. Experiments indicate that Quad calculates the slope more accurately, and from the results of Student's $t$-test, its performance is significantly better than Horn.
机译:在这封信中,提出了一种新算法Quad,以提高坡度精度。根据传统的Horn算法,相邻单元的重要性取决于它们与中心的距离,但是奇怪的是,没有考虑到中心单元本身。自高程的损失会导致无法预测的变形,从而可能导致错误。为了弥补这种不足,Quad算法使用2 $ times $ 2单元邻域模型代替了3 $ times $ 3,因此中心单元自身的高程成为计算过程中最重要的因素。虽然水平和垂直细胞的重要性是玉米细胞的两倍,但中心细胞却比玉米细胞大两倍。 Horn和Quad的比较基于三个不同的人工曲面,在其中我们可以通过推导获得真实的斜率值。实验表明,Quad可以更准确地计算斜率,并且根据Student的$ t $检验结果,其性能明显优于Horn。

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