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

Image reconstruction method and device

An image reconstruction and image technology, applied in the field of communication, can solve problems such as high-frequency information loss

Active Publication Date: 2019-06-28
HUAWEI TECH CO LTD
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, when the super-resolution model trained only based on the pixel mean square error performs image reconstruction, the generated image is obviously smooth and the high-frequency information is seriously lost.

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 reconstruction method and device
  • Image reconstruction method and device
  • Image reconstruction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] First, in order to facilitate understanding of the embodiments of the present application, some concepts or terms involved in the embodiments of the present application are explained.

[0041] (1) Super resolution

[0042] Super resolution (super resolution, SR) refers to the technology of reconstructing low resolution (low resolution, LR) images into high resolution (high resolution, HR) images by computer through image processing methods. A high-resolution image means that the image has a high pixel density, which can provide more image details, which often play a key role in the application.

[0043] Image super-resolution techniques can be divided into two categories: reconstruction-based image super-resolution methods and learning-based image super-resolution methods. Among them, the reconstruction-based image super-resolution method can statistically estimate the high-resolution image with the maximum posterior probability through the frequency domain algorithm o...

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 embodiment of the invention provides an image reconstruction method and device, and the method comprises the steps: inputting a first image into a new super-resolution model to obtain a reconstructed second image, wherein the resolution of the second image is higher than that of the first image, the new super-resolution model is obtained by training an initial super-resolution model by using error loss, the error loss comprises a pixel mean square error and an image feature mean square error, and the image feature comprises at least one of a texture feature, a shape feature, a spatial relationship feature and an image high-level semantic feature. By implementing the embodiment of the invention, the quality of the reconstructed image can be improved.

Description

technical field [0001] Embodiments of the present invention relate to the field of communication technologies, and in particular, to an image reconstruction method and device. Background technique [0002] Image super-resolution reconstruction refers to the technology of reconstructing low-resolution images into high-resolution images using image processing methods, which can effectively improve the clarity of images, and is of great importance to various fields such as video surveillance, camera photography, high-definition television, and medical images. significance. In image super-resolution reconstruction, face image super-resolution reconstruction (face image super-resolution) is widely used, and face super-resolution reconstruction is also called face hallucination. [0003] Currently, methods for face super-resolution reconstruction include methods based on signal reconstruction and methods based on machine learning. Among them, the signal-based reconstruction meth...

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): G06T5/50
CPCG06T5/50G06T7/10G06T3/4053G06T3/4046G06T2207/20081G06T2207/20084G06T5/73
Inventor 遇冰郑传霞冯柏岚
Owner HUAWEI TECH CO LTD