The article proposes the use of PCA transformation carried out with the use of a neural network as a method for compress data obtained from geodetic measurements. In this study, the goal was to assess the effectiveness of the proposed approach, which enables possible to reduce the input data space by determining independent principal components.

The applicability of these methods has been exemplified by the results of vertical displacement measurements of a building. The results of calculations carried out using artificial intelligence assisted and PCA indicates that the approach can be effectively used to compress of geodetic measurement results and then to reproduce them without loss of accuracy of displacement identification. The degree of accuracy of displacement vector reconstruction did not exceed twice the average measurement error, for the most unfavourable situation, amounted to ±0.21 mm. This reduction leads to data compression and makes it possible for the example presented in the article almost to reduce the amount of information stored three times.

这篇文章提出了使用PCA变换和神经网络进行压缩的方法，该方法用于压缩从大地测量获得的数据。在这项研究中，目标是评估所提出方法的有效性，该方法可以通过确定独立的主成分来减少输入数据空间。

这些方法的适用性已通过建筑物的垂直位移测量结果进行了举例说明。使用人工智能辅助和PCA进行的计算结果表明，该方法可以有效地用于压缩大地测量结果，然后重现它们，而不会降低位移识别的准确性。位移矢量重建的精度不超过平均测量误差的两倍，在最不利的情况下，误差为±0.21 mm。这种减少导致数据压缩，并且使本文中的示例几乎可以减少三倍存储的信息量。