A high-resolution hyperspectral computational imaging method, system and medium
A computational imaging and high-resolution technology, applied in the field of high-resolution hyperspectral imaging, can solve problems such as limiting the performance of convolutional neural networks, achieve strong generalization ability, reduce the number of parameters, and improve learning performance
Active Publication Date: 2021-06-22
HUNAN UNIV
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, such methods often ignore the imaging model in spectral super-resolution, which limits the performance of convolutional neural networks.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] Such as figure 1 As described, the high-resolution hyperspectral computational imaging method of this embodiment includes:
[0041]1) Perform spectral upsampling on the input RGB image Y to obtain the initial hyperspectral image X 0 ;
[0042] 2) The initial hyperspectral image X 0 Input to the deep convolutional neural network guided by the pre-trained imaging model, and obtain the hyperspectral image X through iterative solution.
[0043] In this embodiment, the generalized inverse upsampling of the spectral response function is used for the input RGB image Y to obtain the initial hyperspectral image X 0 , perform spectral upsampling on the input RGB image Y to obtain the initial hyperspectral image X 0 The function expression of is:
[0044] (1)
[0045] In the above formula, is the generalized inverse of the spectral response function R.
[0046] Such as figure 1 As shown, the deep convolutional neural network guided by the imaging model in step 2) is ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a high-resolution hyperspectral computing imaging method, system and medium. The method of the invention includes: performing spectral up-sampling on an input RGB image Y to obtain an initial hyperspectral image X 0 ; The initial hyperspectral image X 0 Input to the deep convolutional neural network guided by the pre-trained imaging model, and obtain the hyperspectral image X through iterative solution. The deep convolutional neural network is composed of multiple modules with the same structure. The empirical learning module HPL and the imaging model guidance module IMG are composed, and the hyperspectral prior learning module HPL is used to learn the previous module or the initial hyperspectral image X 0 The prior features of the upsampling results of . The invention can effectively realize the reconstruction from the RGB image to the high-resolution hyperspectral image, and has the advantages of high reconstruction precision, high calculation efficiency, small memory consumption and strong generalization ability.
Description
technical field [0001] The invention relates to high-resolution hyperspectral imaging technology, in particular to a high-resolution hyperspectral computational imaging method, system and medium. Background technique [0002] Hyperspectral imaging technology can obtain image information of dozens or hundreds of spectral bands at the same time, and rich spectral information is helpful for accurate identification of substances in the scene. Therefore, hyperspectral imaging technology is widely used in earth observation, military monitoring, Environmental monitoring, geological exploration, medical detection and face recognition and other fields. However, due to the limitations of optical imaging systems, it is difficult for existing optical imaging systems to directly acquire high-resolution hyperspectral images. At the same time, spectral imaging equipment is expensive, which greatly limits the application of hyperspectral images. On the other hand, existing imaging systems...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T3/40G06N3/04
CPCG06T3/4076G06T2207/20081G06N3/045G06T3/4046G06T3/4053
Inventor 李树涛佃仁伟郭安静康旭东孙斌方乐缘卢婷
Owner HUNAN UNIV