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

Image super-resolution reconstruction method and device

A super-resolution reconstruction and image technology, applied in the field of image processing, can solve problems such as poor quality of high-resolution images and affect user experience, and achieve the effects of accelerating image feature propagation, reducing complexity, and reducing the number

Pending Publication Date: 2019-08-16
HUAWEI TECH CO LTD
View PDF4 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the quality of high-resolution images obtained by this method is poor, which affects user experience

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 and device
  • Image super-resolution reconstruction method and device
  • Image super-resolution reconstruction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0076] The technical solution in this application will be described below with reference to the accompanying drawings.

[0077] For ease of understanding, the neural network is first introduced in detail. A neural network generally includes multiple neural network layers, and each neural network layer can implement different calculations or operations. Common neural network layers include convolution layers, pooling layers, and full-connection layers.

[0078] figure 1 It is the basic frame diagram of convolutional neural networks (CNN). see figure 1 , the convolutional neural network includes convolutional layers, pooling layers, and fully connected layers. Wherein, multiple convolutional layers and multiple pooling layers are arranged alternately, and the convolutional layer may be followed by a convolutional layer or a pooling layer.

[0079] The convolutional layer is mainly used to perform convolution operation on the input matrix, and the pooling layer is mainly use...

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 provides an image super-resolution reconstruction method and device. The image super-resolution reconstruction method comprises the steps: dividing a to-be-reconstructed image into n image blocks, and enabling n to be an integer greater than or equal to 2; determining the type of each image block in the n image blocks; according to the type of each image block in the n image blocks,obtaining n image blocks, and reconstructing each image block through a convolutional neural network reconstruction model; obtaining reconstructed image blocks of each image block in the n image blocks, wherein the convolutional neural network reconstruction model comprises a plurality of groups of reconstruction parameter sets, and reconstruction parameter sets in a convolutional neural network reconstruction model used during reconstruction of different types of image blocks are different; and splicing the reconstructed image blocks of the n image blocks together to obtain a super-resolutionreconstructed image of the image. According to the embodiment of the invention, the high-resolution image quality can be improved, and the user experience is improved.

Description

technical field [0001] The present application relates to the field of image processing, and more specifically, to a method and device for image super-resolution reconstruction. Background technique [0002] In the field of electronic image applications, people often expect to obtain high resolution (high resolution, HR) images. However, limited by the high cost of imaging equipment or constraints of physical conditions, users may only be able to obtain low-resolution images. Therefore, in order to obtain high-resolution images, a technology is needed that can High-resolution images are processed to obtain desired high-resolution images. [0003] Image super-resolution (SR) reconstruction technology is to reconstruct and upgrade low-resolution (LR) images (such as 1080P: 1920x1080) to HR images (such as 4K: 3840x2160) through certain algorithms. [0004] High-resolution images have higher pixel density, more detailed information, and more delicate image quality, so image s...

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): G06T3/40G06N3/04G06N3/08
CPCG06T3/4046G06T3/4053G06N3/084G06N3/045
Inventor 徐威赵荣乾
Owner HUAWEI TECH CO LTD