Unlock instant, AI-driven research and patent intelligence for your innovation.

A Multispectral Image Inversion Method Based on Residual Learning Convolutional Neural Network

A convolutional neural network, multi-spectral image technology, applied in the field of multi-spectral imaging, can solve problems such as difficulty in obtaining clear images, and achieve the effect of improving speed

Active Publication Date: 2022-03-18
ZHONGBEI UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the deficiencies in the prior art, the present invention provides a multispectral image inversion method based on residual learning convolutional neural network, which solves the problem of difficult acquisition of clear images common in existing multispectral imaging systems

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
  • A Multispectral Image Inversion Method Based on Residual Learning Convolutional Neural Network
  • A Multispectral Image Inversion Method Based on Residual Learning Convolutional Neural Network
  • A Multispectral Image Inversion Method Based on Residual Learning Convolutional Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] Next, the technical solutions in the embodiments of the present invention will be described in connection with the drawings of the embodiments of the present invention, and it is understood that the described embodiments are merely the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art are in the range of the present invention without making creative labor premise.

[0031] Such as Figure 1-2 As shown, a multi-spectral image inversion method based on residual learning convolutional neural network, including the following steps:

[0032] S1, establish the main channel and the slave channel image library. The main channel image and the main channel image of k target objects are collected by the multi-spectral imaging system, respectively, wherein the multi-spectral imaging system includes n channels, any selection of the 第 {{1, 2, ..., n} chan...

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 present invention relates to the field of image inversion problems in the field of multispectral imaging, and more specifically, relates to a multispectral image inversion method based on residual learning convolutional neural network. This method first selects one channel as the master channel (other channels as slave channels), uses traditional methods (focusing or image deblurring algorithms) to obtain a clear image of the master channel, and then inputs the clear image to each slave channel residual The neural network model can calculate the output of each slave channel residual neural network model (that is, the residual between the slave channel image and the main channel image), and finally add the residual to the main channel image to invert each Clear image from channel.

Description

Technical field [0001] The present invention relates to the field of image inversion in the field of multi-spectral imaging, and more particularly, a multi-spectral image inversion method based on residual learning convolutional neural network. Background technique [0002] Existing studies have found that the imaging quality of the multi-spectral imaging system is usually poor, and it is difficult to obtain a clear image of each channel. The main reason is caused by factors such as out-of-focus, diffraction direction wavelength. In response to this problem, many researchers at home and abroad have launched in-depth research. For defocus-blurring problems, early studies have passed the imaging blur of color difference by focusing, mainly calculating image sharpness through a certain algorithm and using a stepper motor to focus on the front optical lens, but imaging in wide spectral range In the measurement, the time consumption is complete, and the measurement effect is long, and...

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): G06T5/00G06T5/50G06N3/04
CPCG06T5/50G06T2207/20081G06T2207/10036G06N3/045G06T5/00
Inventor 陈媛媛李墅娜常晓丽景宁王志斌
Owner ZHONGBEI UNIV