Multi-camera hyperspectral imaging method and system based on deep learning, and medium

A hyperspectral imaging, multi-camera technology, applied in machine learning, color/spectral characteristic measurement, instruments, etc., can solve the problems of difficult hyperspectral imaging, limited application scope, inconvenient portability, etc., to reduce the time of hyperspectral imaging , huge practical value, the effect of improving convenience

CN111579506AActive Publication Date: 2020-08-25HUNAN UNIV
1 Cites 6 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2020-08-25

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a multi-camera hyperspectral imaging method and system based on deep learning, and a medium. The multi-camera hyperspectral imaging method provided by the invention comprises the following steps of performing hyperspectral imaging on multiple cameras; constructing a color back projection network module according to spectral response characteristics of the color cameras; constructing a gray scale back projection network module according to the spectral response characteristics of the gray scale camera, constructing an iterative back projection hyperspectral reconstruction network by using the color back projection network module and the gray scale back projection network module, and respectively obtaining a color image and a gray scale image obtained by the color cameras and the gray scale camera for the same imaging target; and inputting the color image and the grayscale image into the trained iterative back projection hyperspectral reconstruction network to obtain a hyperspectral image of an imaging target. The method can effectively guarantee the imaging quality, greatly shortens the hyperspectral imaging time, does not need an additional auxiliary deviceor hardware, greatly improves convenience of hyperspectral imaging, and is wide in application range.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present invention relates to a hyperspectral imaging method, in particular to a multi-camera hyperspectral imaging method, system and medium based on deep learning. Use product authenticity identification. Background technique

[0002] Hyperspectral images usually contain dozens or hundreds of spectral bands, and have rich and differential spectral information. They have been widely used in remote sensing ground feature classification, medical aided diagnosis, agricultural pest identification, agricultural product drug residue detection, and authenticity identification of daily products. and other fields. However, limited by imaging components and sensor technology, current hyperspectral imaging devices often have disadvantages such as slow imaging time, large volume, and inconvenient portability. However, existing portable hyperspectral imaging solutions, such as multi-lens multispectral imaging systems, mobile phone electric rotating multispect...

Examples

Embodiment Construction

[0047] The following will take a smartphone as a typical device with multiple cameras as an example to further describe the deep learning-based multi-camera hyperspectral imaging method, system and medium of the present invention in detail. Undoubtedly, the realization of the deep learning-based multi-camera hyperspectral imaging method, system and medium of the present invention is not limited to the specific collection and processing device of the smart phone, as long as it is a computing device with a color camera or a grayscale camera. accomplish. In order to make it easier for those skilled in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings. It should be understood that the embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0048] Such as figure 1 As shown, th...