Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Multispectral image super-resolution reconstruction method based on color image fusion

A technology for super-resolution reconstruction and multi-spectral images, applied in the field of improving the spatial resolution of multi-spectral images, can solve the problems of insufficient fusion information, low reconstruction accuracy, inappropriate fusion methods, etc. Noise ratio, application scenarios require loose effects

Active Publication Date: 2020-08-14
ZHEJIANG UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The present invention aims at problems such as insufficient fusion information and inappropriate fusion methods leading to low reconstruction accuracy in existing multispectral super-resolution imaging methods, and proposes a multispectral image super-resolution reconstruction method using RGB image boundary guidance and content fusion method

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multispectral image super-resolution reconstruction method based on color image fusion
  • Multispectral image super-resolution reconstruction method based on color image fusion
  • Multispectral image super-resolution reconstruction method based on color image fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] In the following, the method of the present invention is used to realize super-resolution reconstruction of multispectral images. In order to compare the reconstruction accuracy of the boundary structure, the simulation data is first used to generate high-resolution RGB images and low-resolution multispectral images, and then the method of the present invention is used for super-resolution reconstruction, and the reconstruction results are compared with the known real image Compare. From image 3 It can be seen from the figure that due to the large noise of the real image, the multispectral image obtained by using the dictionary learning and sparse representation algorithm is blurred, and its boundary structure almost disappears; while the multispectral image obtained by the algorithm of the present invention is clear, and due to the use of Guided by the boundary information of the RGB image, the reconstructed boundary structure is well preserved.

Embodiment 2

[0050] The reconstruction result of using the method of the present invention to improve the spatial resolution of a multispectral image is measured below from a quantitative perspective. From Figure 4 It can be seen in (a) according to the measurement index that the smaller the root mean square error value is, the higher the spatial domain reconstruction accuracy is, the estimation results obtained by using the algorithm based on dictionary learning and sparse representation or the algorithm based on spectral clustering regularization items are in the spatial The quality on the domain is poor, (b) according to the smaller the spectral angle mapping value, the higher the spectral domain reconstruction accuracy is, the estimated result is obtained by using the algorithm based on dictionary learning and sparse representation or the algorithm based on spectral clustering regularization terms The quality in the spectral domain is also poor; in contrast, the reconstruction accurac...

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

No PUM Login to View More

Abstract

The invention discloses a multi-spectral image super-resolution reconstruction method based on color image fusion. Firstly, the high-resolution color image and low-resolution multi-spectral image areacquired and registered, then the inverse camera response function and spectral sensitivity function coupled in the color image are acquired, and the observation model based on the content of the acquired image is constructed. High-resolution multispectral images are solved by this model, Firstly, the boundary structure information is extracted from the captured RGB image, In order to guide the reconstruction of high-resolution multispectral images, the basis of the solution space is trained from the acquired multispectral images, and an iterative algorithm framework based on image fusion is constructed to solve the coefficients of the multispectral images on the spatial basis. Finally, the coefficients are combined with the spatial basis to obtain the high-resolution multispectral images.The invention reduces the error of the traditional multi-spectral image super-resolution method due to the information loss and improves the reconstruction precision of the multi-spectral image by utilizing the boundary guidance and the content fusion of the RGB image.

Description

technical field [0001] The invention relates to improving the spatial resolution of multi-spectral images, in particular to a method for super-resolution reconstruction of multi-spectral images by using boundary guidance and content fusion of color images. Background technique [0002] Multispectral imaging can record rich spectral information of a scene, so it has attracted extensive attention in various fields, such as biology, remote sensing, color reproduction, etc. A multispectral imaging system usually consists of a tunable filter bank and a monochromatic camera, which can acquire a series of continuous narrow-band channel images in the visible spectral band. Multispectral imaging systems can achieve very high spectral domain resolution, but due to system hardware and other factors, it has serious limitations in spatial domain resolution. [0003] In order to improve the spatial resolution of multispectral images, a widely used method in the field of remote sensing is...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T3/40
CPCG06T3/4053
Inventor 沈会良潘之玮
Owner ZHEJIANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products