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

Multi-focus image fusion method based on zero sample learning

A multi-focus image and sample learning technology, applied in neural learning methods, image enhancement, image analysis, etc., can solve problems such as estimating point spread function and depth ill-posed

Active Publication Date: 2021-08-27
HARBIN INST OF TECH
View PDF8 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] These supervised learning-based models using synthetic datasets may differ from the real imaging process, which needs to consider the point spread function (PSF) and the distance between the object and the lens, but estimating PSF and depth are both is a very ill-posed problem

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
  • Multi-focus image fusion method based on zero sample learning
  • Multi-focus image fusion method based on zero sample learning
  • Multi-focus image fusion method based on zero sample learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them; based on this The embodiments in the invention, and all other embodiments obtained by persons of ordinary skill in the art without creative efforts, all belong to the scope of protection of the present invention.

[0049] Because the untrained network can be used as a prior for many low-level vision tasks without any training data, a multi-focus image fusion network IM-Net can be used. IM-Net consists of two joint sub-networks I-Net and M -Net, I-Net models the depth prior of the fusion image, and M-Net models the depth prior of the focused image, and realizes zero-shot learning through the extracted prior information.

[0050] figure 1 The main structure ...

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 a multi-focus image fusion method based on zero sample learning. Information contained in an input multi-focus image is fused by using a multi-focus image fusion network structure IM-Net, the IM-Net comprises two combined sub-networks I-Net and M-Net, depth prior modeling is performed on a fused image by the I-Net, and depth prior modeling is performed on a focus image by the M-Net. Zero sample learning is realized through extracted prior information, reconstruction constraint is applied to IM-Net so as to ensure that information of a source image pair can be better transmitted to a fused image, high-level semantic information can keep brightness consistency of adjacent pixels, guide loss provides guidance information for the IM-Net to search for a clear area, and experimental results show that the method is effective.

Description

technical field [0001] The invention belongs to multi-focus image fusion, and in particular relates to a multi-focus image fusion method based on zero-sample learning. Background technique [0002] In the imaging system, due to the limitation of the depth of field (DOF), objects outside the focal plane of the camera will become blurred, and it is difficult to obtain a fully focused image, which will lead to a significant decline in imaging quality. In recent years, researchers have proposed various multi-focus image fusion (MFIF) algorithms to solve this problem. Multi-focus image fusion algorithms can fuse the focal regions of source images with different focal lengths in the same scene to obtain high-quality fully-focused images, with a wide range of applications including digital photography, microscopic imaging and advanced vision tasks or simply to obtain more Good visual perception. [0003] In recent years, impressive progress has been made in the problem of multi-f...

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 Applications(China)
IPC IPC(8): G06T5/50G06N3/04G06N3/08
CPCG06T5/50G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/20221G06N3/045
Inventor 江俊君胡星宇刘贤明马佳义
Owner HARBIN INST OF TECH