トンネル(5)

[III-227] Examination of correction method of spectrum change by light source for rock type classification method using hyperspectral data for on-site application

〇SIMIN ZHAI¹, Sadatoshi Ohmori¹, Narihiro Owada², Youhei Wakamura² (1.Penta-ocean construction co..Ltd, 2.Akita University)

Keywords:Tunnel face evaluation, Hyperspectral, Light source

Geological evaluation of the tunnel face in tunnel construction is important for optimal design and construction. A hyperspectral (HS) camera is a useful tool for mechanically grasping geological conditions, and some experiments have shown that the data successfully identified igneous rocks through deep learning. On the other hand, the spectral data of the HS camera depends on the spectral characteristics of the light source, it is difficult to stabilize the light source environment at the construction site. This article reports a method for correcting spectral data, even when taking hyperspectral images under different light source conditions.