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

Image super-resolution reconstruction method based on sparse self-encoding network and speed learning

A sparse auto-encoding and extreme-speed learning technology, applied in the field of image super-resolution reconstruction based on sparse auto-encoding network and extreme learning, can solve the problems of not finding neighbors, inaccurate reconstruction, and average reconstructed image quality, so as to improve the accuracy , the effect of enhancing the versatility

Active Publication Date: 2017-11-24
SUZHOU WENJIE SENSING TECH
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the feature extraction of these methods is not accurate enough, and accurate neighbors are often not found, and the feature extraction and reconstruction steps are performed independently, resulting in inaccurate reconstruction, and the quality of the reconstructed image is average.

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 sparse self-encoding network and speed learning
  • Image super-resolution reconstruction method based on sparse self-encoding network and speed learning
  • Image super-resolution reconstruction method based on sparse self-encoding network and speed learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] The present invention is an image super-resolution reconstruction method based on sparse autoencoder network and extremely fast learning, see figure 1 , the present invention includes the following steps to image super-resolution reconstruction:

[0033] Step 1: Input the training sample image pair, and divide the low-resolution training sample image into the size of The low-resolution training example image patches of and pulled into a vector Segment the high-resolution training example image into patches with the low-resolution training example image in the same way The corresponding size is High-resolution training example image patches of and pulled into a vector Where n represents the dimension of the low-resolution training sample image block vector, N represents the scale of high-resolution and low-resolution training sample image blocks, and s represents the upsampling multiple;

[0034] Step 2: Convert low-resolution training example image patch vector...

Embodiment 2

[0044] The image super-resolution reconstruction method based on sparse autoencoder network and extremely fast learning is the same as that in Embodiment 1, refer to the appendix figure 1 , the specific steps of the image super-resolution reconstruction method based on sparse autoencoder network and extremely fast learning of the present invention include:

[0045] 1. An image super-resolution reconstruction method based on representation learning and neighborhood constraint embedding (an image super-resolution reconstruction method based on sparse autoencoder network and extremely fast learning), characterized in that it includes the following steps:

[0046] Step 1: Input training sample image pairs, see figure 2 , image 3 , split the low-resolution training sample image into a size of The low-resolution training example image patches of and pulled into a vector Segment the high-resolution training example image into patches with the low-resolution training example im...

Embodiment 3

[0058] The image super-resolution reconstruction method based on sparse autoencoder network and extremely fast learning is the same as that in Embodiment 1-2.

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 sparse self-encoding network and speed learning, and a problem of inaccurate recovery caused by inaccurate characteristic extraction and independent characteristic extraction and reconstruction steps is solved. The method comprises a step of inputting a low resolution test image, partitioning the inputted low resolution test image through an overlapping mode, and constructing low resolution test image blocks, a step of allowing each low resolution test image block to pass a hierarchical sparse self-encoder to extract sparse characteristics, a step of allowing the sparse characteristics to pass a speed learning machine obtained through prior learning to be projected to a high resolution pixel space to obtain high resolution image blocks, and a step of aggregating the high resolution image blocks to obtain a final high resolution image to complete the image super-resolution reconstruction. According to the method, the multi-layer sparse self-encoding network is used to learn the sparse characteristics, the characteristic extraction and reconstruction are integrated in a unified network framework, the quality of a reconstructed image is effectively improved and the detail information is improved, and the method is suitable for the image super-resolution reconstruction of various natural images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and mainly relates to image super-resolution reconstruction, in particular to an image super-resolution reconstruction method based on sparse self-encoding network and extremely fast learning, which can be used for super-resolution reconstruction of various natural images and character images . Background technique [0002] Image super-resolution reconstruction technology can restore high-resolution images from low-resolution images by means of signal processing, and has been widely used in many fields. Traditional reconstruction-based super-resolution methods mostly use prior knowledge construction constraints such as image edge characteristics, pixel non-negativity, and local smoothness characteristics, and the prior information of the image itself is insufficiently utilized. The resulting image recovered when it is large is too smooth. Freeman et al. proposed a learning-based method...

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
CPCG06T3/4076G06F18/214
Inventor 王敏王勇
Owner SUZHOU WENJIE SENSING TECH