摘要
作为高效内插技术,DEM曲面内插迅速崛起于测绘科学领域,如何高精度、高效率地内插成为当前研究的热点。本次研究选取辽宁省调兵山实验区域为研究对象,利用多项式曲面内插法结合BP神经网络算法进行实验,主要研究各方法之间的精度和BP神经网络对其优化的效果。研究结果表明:1)实验区域内使用单种内插方法内插时,双三次多项式内插法的内插精度最佳,双四次最差;2)BP神经网络能较好地优化双多次多项式曲面内插精度;3)BP双三次多项式曲面内插法是一种适合调兵山地区的内插方法。
As a highly efficient interpolation technique,DEM surface interpolation is rapidly emerging in the field of Surveying and mapping. How to insert the high-precision and high-efficiency into the hot spot of the current era. This study selected Diaobingshan experimental area in Liaoning Province as the research object,using polynomial interpolation method combined with BP neural network algorithm experiment,mainly to study various methods and accuracy of BP neural network on the optimization effect. The results show that：（1）the experimental region using a single interpolation method of interpolation,double three polynomial interpolation accuracy is best,four is the worst;（2）the BP neural network can better optimize the double times polynomial surface interpolation accuracy;（3） BP double three times polynomial interpolation method is a kind of interpolation method which is suitable for Diaobingshan area.
作者
范强
朱添翼
FAN Qiang;ZHU Tianyi(College of Surveying,Mapping and Geographical Science,Liaoning Technical University,Fuxin 123000,China)
出处
《测绘与空间地理信息》
2018年第7期11-14,共4页
Geomatics & Spatial Information Technology
