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

Tandem type single-image super-resolution reconstruction method

A super-resolution reconstruction and super-resolution technology, which is applied in the field of serial single-image super-resolution reconstruction to achieve good reconstruction effect

Pending Publication Date: 2020-07-10
CHINA JILIANG UNIV
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, the super-resolution image reconstruction method of the existing technology is still difficult to meet the needs of actual use, and there is room for further improvement

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
  • Tandem type single-image super-resolution reconstruction method
  • Tandem type single-image super-resolution reconstruction method
  • Tandem type single-image super-resolution reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] The present invention is based on first training filters and mappings, then using the trained filters to perform convolutional sparse coding reconstruction, and then using pre-stored anchor points and corresponding mappings to perform improved fixed neighborhood regression super-resolution reconstruction. Finally, the self-sample anchor neighborhood regression is further used for image enhancement.

[0060] Basic train of thought of the present invention is:

[0061] First, the low-resolution and high-resolution filters and the mapping of low-resolution feature maps to high-resolution feature maps are obtained through convolutional sparse coding training.

[0062] Secondly, use the filter and mapping obtained from the training to reconstruct the low-resolution training set as a new low-resolution training set, and form a training pair with the high-resolution training set for dictionary training of improved fixed neighborhood regression, and each Anchor points and the ...

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 tandem type single-image super-resolution reconstruction method. The method comprises the steps: taking a whole ground resolution image as input convolutional sparse codes for reconstruction; employing an improved fixed neighborhood regression algorithm taking an image block as input for reconstruction; finally, in order to introduce image internal statistical information, establishing a self-sample pyramid, and taking the self-sample pyramid as a training set to perform reconstruction by using self-sample fixed domain regression. According to the invention, the advantages of each layer of method can be utilized to reconstruct the image for multiple times, and a better reconstruction effect can be achieved.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a serial single image super-resolution reconstruction method. Background technique [0002] Single-image super-resolution reconstruction reconstructs a high-resolution image by recovering high-frequency details and removing blur. Obviously, this single image super-resolution reconstruction algorithm has important applications in many fields, such as surveillance equipment, satellite images, medical images, etc. But single-image super-resolution is a severely ill-posed problem, since the observed low-resolution image is much smaller than the desired high-resolution image, so the reconstruction of high-resolution images is uncertain. Various approaches have been proposed to address this dilemma: [0003] Most of the proposed methods can be classified into 3 categories: interpolation-based methods, reconstruction-based methods and learning-based methods. [0...

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
IPC IPC(8): G06T3/40G06K9/62
CPCG06T3/4053G06F18/28
Inventor 曹飞龙张焯林
Owner CHINA JILIANG UNIV