Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for imaging object passing behind scattering medium based on convolutional neural network

A convolutional neural network and medium technology, applied in the field of machine learning and image reconstruction, can solve the problem of high structural complexity, and achieve the effect of enhancing the ability to extract local feature details

Inactive Publication Date: 2020-10-02
NANJING UNIV OF SCI & TECH
View PDF5 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, the structural complexity of real objects is very high, requiring the reconstruction capability to be powerful enough to retrieve images of hidden objects

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
  • Method for imaging object passing behind scattering medium based on convolutional neural network
  • Method for imaging object passing behind scattering medium based on convolutional neural network
  • Method for imaging object passing behind scattering medium based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] Below in conjunction with accompanying drawing and specific embodiment, further illustrate the present invention, should be understood that these embodiments are only for illustrating the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various aspects of the present invention Modifications in equivalent forms all fall within the scope defined by the appended claims of this application.

[0047] Such as figure 1 As shown, the present invention is based on a convolutional neural network for imaging an object after a scattering medium, specifically comprising the following steps:

[0048] Step 1. Design the encoder-decoder structure, the encoder-decoder structure is connected by skipping layers to enhance the ability to extract local features and details;

[0049] Step 2. Design the convolutional neural network PDSNet, the structure of the convolutional ne...

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 a method for imaging an object behind a scattering medium based on a convolutional neural network PDSNet. According to the method, the traditional speckle correlation imaging algorithm principles are combined, the design and optimization of the network are guided, and the limitation of the optical memory effect OME on the imaging field angle FOV is eliminated in a data driving mode. The convolutional neural network PDSNet is a neural network structure suitable for a random scale and a complex target. The hidden object recovery capability of the convolutional neural network PDSNet is experimentally tested, and at least 40 times of optical memory effect range expansion is realized on the premise that the average PSNR is kept above 24dB. And meanwhile, under an untrained scale, the average PSNR of the recovered image is more than 22dB, and complex targets such as a human face are successfully reconstructed. Experimental results given in the invention verify the accuracy and effectiveness of the method.

Description

technical field [0001] The invention relates to a method for imaging an object passing through a scattering medium based on a convolutional neural network, which belongs to the field of machine learning and image reconstruction. Background technique [0002] Scattering media are ubiquitous in biological tissues and are the main source of interference in the field of astronomical imaging. Many new imaging methods have been proposed to achieve imaging through severely disordered media. Typical methods include optical coherence tomography, wavefront modulation, transfer matrix based and point spread function based image reconstruction algorithms. Katz et al proposed a speckle correlation imaging technique based on the optical memory effect. Unlike the above methods, this method can be imaged through strongly scattering media, can be applied to dynamic scattering media, and does not require an additional reference source. [0003] Due to the presence of OME, scattering media ...

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): G06T11/00G06N3/04G06N3/08G06K9/48G06K9/62
CPCG06T11/003G06N3/084G06V10/476G06V10/46G06N3/045G06F18/25
Inventor 韩静柏连发张毅赵壮孙岩郭恩来朱硕师瑛杰顾杰崔倩莹戚浩存左苇吕嫩晴
Owner NANJING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com