LAUSR

The principle of hyperspectral imaging leads to a variety of stripe noise in hyperspectral images, especially the wide-stripe noise, which brings great obstacles to the interpretation and application of hyperspectral images. Aiming at the wide-stripe noise of hyperspectral images of two-level production data, considering from the effect of filtering noise and the ability of protecting detail, this paper proposed a fused wide-stripe removal method based on the wavelet transform and local interpolation (WTLI), called the WTLI algorithm. On one hand, it uses the wavelet transform to remove the stripe noise as much as possible; on the other hand, it uses the local interpolation to protect more geometric and detailed information, so as to achieve the purpose of removing noise and protecting the useful information. A series of comparative experiments were carried out with hyperspectral image data. Not only have good experimental results been obtained, but also this shows that the WTLI algorithm has better stability and universality.