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

Image Super-resolution Reconstruction Method Based on Genetic Algorithm and Regularized Prior Model

A genetic algorithm and a priori model technology, applied in the field of image super-resolution reconstruction, can solve the problems that affect the image reconstruction effect and the high randomness of the high-resolution image to be estimated, and achieve the effect of improving the reconstruction effect.

Active Publication Date: 2017-05-31
XIDIAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In recent years, there have also been methods of using genetic algorithms to reconstruct super-resolution images, but because there is no guidance from any texture and edge information of the image, the randomness of the obtained high-resolution image to be estimated is too large, which affects the reconstruction effect of the image

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
  • Image Super-resolution Reconstruction Method Based on Genetic Algorithm and Regularized Prior Model
  • Image Super-resolution Reconstruction Method Based on Genetic Algorithm and Regularized Prior Model
  • Image Super-resolution Reconstruction Method Based on Genetic Algorithm and Regularized Prior Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] Combine below figure 1 The specific implementation steps of the present invention will be further described in detail.

[0048] Step 1: Learn to acquire 200 sub-dictionaries from natural images.

[0049] Input 5 natural images, from these 5 natural images get image blocks containing a lot of edge and structure information, use K-means to divide these image blocks into 200 categories, and use PCA to obtain a sub-dictionary from each category Φ k .

[0050] Step 2: Obtain a high-resolution image Xs, and extract the initial estimate X of the brightness component of Xs.

[0051] (2a) Input the low-resolution image LR, and use bicubic interpolation to enlarge it by 3 times to obtain the initial estimate Xs of the high-resolution image;

[0052] (2b) Convert the initial estimate Xs of the high-resolution image from the red, green, and blue RGB space to the YCbCr color space to obtain the initial estimate of the luminance component Y, the blue chrominance component estimate Cb, and th...

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 an image super-resolution reconstruction method based on a genetic algorithm and a regular prior model, and mainly solves a problem of poor quality of a reconstruction result of a traditional method. The image super-resolution reconstruction method comprises the following implementation steps: (1) learning one group of sub-dictionaries from a natural image; (2) obtaining the luminance component estimation X of a high-resolution image Xs after a low-resolution (LR) image is magnified through interpolation by three times; (3) constructing an initial population; (4) calculating the fitness value of each individual; (5) selecting and copying the individuals in a parent population; (6) successively carrying out cross and variation to the individuals of the parent population; (7) repeating the steps (5) and (6) for twenty times to obtain an optimal solution X'; (8) carrying out local optimization on the X' by utilizing the regular prior model; and (9) repeating steps (3) to (8) for four times to obtain a luminance component X2 of a high-resolution image, and finally combining the high-resolution image. Image edges and texture information can be favorably kept, and the image super-resolution reconstruction method can be used for image identification and target classification.

Description

Technical field [0001] The invention belongs to the technical field of image processing, and specifically is an image super-resolution reconstruction method used for super-resolution reconstruction of various natural images. Background technique [0002] In practical applications, it is not easy to directly obtain high-quality images with the required resolution due to the limitation of imaging equipment capabilities and the influence of complex working environment conditions. The image super-resolution reconstruction SR technology breaks through the limitations of imaging equipment and the environment, recovers the high-frequency information lost when the image is acquired from the low-resolution observation image, and reconstructs a high-resolution image. Therefore, SR has very important application prospects in the fields of video, remote sensing, medicine and security monitoring, and has always been a hot spot in image science research and its engineering applications. [0003...

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): G06T5/50G06T3/40G06N3/12
Inventor 李阳阳焦李成李亚肖马文萍尚荣华马晶晶杨淑媛侯彪
Owner XIDIAN UNIV