PROJECT TITLE :

A Manifold Alignment Approach for Hyperspectral Image Visualization With Natural Color

ABSTRACT:

The trichromatic visualization of tons of bands in a very hyperspectral image (HSI) has been a full of life research topic. The visualized image shall convey as abundant info as doable from the original information and facilitate easy image interpretation. However, most existing methods show HSIs in false color, which contradicts with user experience and expectation. In this paper, we have a tendency to propose a new framework for visualizing an HSI with natural color by the fusion of an HSI and a high-resolution color image via manifold alignment. Manifold alignment projects many knowledge sets to a shared embedding space where the matching points between them are pairwise aligned. The embedding house bridges the gap between the high-dimensional spectral house of the HSI and also the RGB space of the colour image, making it possible to transfer natural color and spatial information in the color image to the HSI. During this way, a visualized image with natural color distribution and fine spatial details can be generated. Another advantage of the proposed methodology is its flexible information setting for varied scenarios. As our approach only needs to go looking a limited range of matching pixel pairs that gift the identical object, the HSI and the color image can be captured from the identical or semantically similar sites. Moreover, the learned projection function from the hyperspectral data house to the RGB area will be directly applied to different HSIs acquired by the same sensor to realize a fast overview. Our method is also in a position to visualize user-specified bands as natural color pictures, which is terribly useful for users to scan bands of interest.