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

Image deblurring method based on Scale-Encoder-Decoder-Net network

A deblurring and image technology, applied in biological neural network models, image enhancement, image analysis, etc., to achieve good image detail recovery, good image detail recovery, and reduced recovery time.

Pending Publication Date: 2020-12-18
NORTHWESTERN POLYTECHNICAL UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] In order to avoid the shortcomings of the existing technology, the present invention proposes an image deblurring method based on the Scale-Encoder-Decoder-Net network, aiming at solving the shortcomings of the existing deep learning deblurring algorithm, by comparing the existing optimal algorithm, the present invention improves the time complexity by 0.12s, and improves the image restoration performance by an average of 0.8dB

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 deblurring method based on Scale-Encoder-Decoder-Net network
  • Image deblurring method based on Scale-Encoder-Decoder-Net network
  • Image deblurring method based on Scale-Encoder-Decoder-Net network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0051] figure 1 is the image deblurring mechanism of the present invention. Among them, the blurred images of different scales are restored to deblurred images of different scales through the Scale-Encoder-Decoder-Net network of different scale layers, the global feature information is restored at a high scale, and the local feature information is restored at a low scale.

[0052] The concrete steps of the embodiment of the present invention are as follows:

[0053] (1) Design the Scale-Encoder-Decoder-Net network, the principle is as follows figure 2 shown. The blurred image is deblurred through the Scale-Encoder-Decoder-Net network.

[0054] (2) Use the following loss function to train the network,

[0055]

[0056] Among them, L represents the total loss function of multi-scale, I i and Represents the clear image and restored image of the i-th scale, N i Represents the number of image pixels at the i-th scale. Finally, the loss between the restored image 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 belongs to the field of digital image processing, and relates to an image deblurring method based on Scale-Encoder-Decoder-Net. According to the method, blurred image restoration is realized by using a Scale-Encoder-Decoder-Net network, so that the problems of long restoration time, poor image detail restoration effect and the like of an existing deep learning deblurring algorithm are solved. The invention provides a Scale-Encoder-Decoder-Net network, and aims to solve the defects of an existing deep learning deblurring algorithm, and compared with an existing optimal algorithm,the method has the advantages that the restoration time is prolonged by 0.12 s, and the performance of restored images is averagely improved by 0.8 dB.

Description

technical field [0001] The invention belongs to the field of digital image processing, and relates to an image deblurring method based on a Scale-Encoder-Decoder-Net network. Background technique [0002] Today, smartphones, cameras and other photographic equipment are widely used, taking pictures has become an important way for people to freeze moments and record their lives. When shooting hand-held, whether using a mobile phone, a card machine, or a professional SLR camera, it is easy to produce different degrees of blur due to camera shake, which seriously affects the imaging quality of the image in a certain sense and brings poor visual perception. . [0003] Image deblurring technology is an important technology of image preprocessing in image and video processing. On data with a high degree of ambiguity, the accuracy of artificial intelligence algorithms such as face recognition, license plate recognition, vehicle recognition, and pedestrian detection will drop signi...

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): G06T5/00G06N3/04G06N3/08
CPCG06N3/08G06T2207/10004G06N3/045G06T5/73
Inventor 杨宁秦毅杰郭雷
Owner NORTHWESTERN POLYTECHNICAL UNIV