Hyper-spectral imaging provides the possibility of describing specific characteristics of materials in air, land and water depending on reflections of each material in a specific zone of the electromagnetic spectrum. Hyper-spectral sensors capture images in continuous and narrow spectral bands in the visible, near-infrared, and mid-infrared spectral regions.
These sensors look at objects using a large part of the electromagnetic spectrum. Some objects have unique characteristics in certain electromagnetic spectrums. These spectral signatures, known as "spectral signatures," help to identify materials in images. But the main problem with these sensors is that the images they get in bad weather conditions and in the presence of dust or rain or snow are of very poor quality. Therefore, we go to SAR images, which are able to form a high-resolution image even in bad weather conditions.
The method proposed in this paper is to obtain the final elements of hyper-spectral images using the decomposition method into a multilayer negative matrix. The reason for using this method is that, unlike the high-spectral images that form the image in a number of high-traffic zones, we can create the image in 8 different frequencies, so the most important thing is to get all the hidden information at those frequencies.
And also, we compare between the RDA, SENMAV, OMP, ISMA and ADMM algorithms.