Hyperspectral imaging collects data over hundreds of contiguous narrow bands, a single data cube is far larger in volume than an ordinary RGB image, and adjacent bands are highly correlated; in real scenarios there is no noise-free hyperspectral reference, so supervision signals can usually only be constructed by synthetic noise or physical simulation.