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

Fresnel hologram frequency domain gating filtering method based on deep learning

A technology of Fresnel holography and deep learning, applied in the field of Fresnel hologram frequency-domain gating filtering, can solve the problems of poor image reconstruction effect, complex compression algorithm, and high application requirements, so as to improve effective utilization and reduce space occupation Compared with the effect of improving real-time performance

Pending Publication Date: 2021-12-24
KUNMING UNIV OF SCI & TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problems of insufficient compression rate of holograms, complex compression algorithms, difficulties in real-time transmission and storage, poor image reconstruction effect, and high requirements for the application of holograms in current true-color scenes, and propose a Fresnel based on deep learning. Frequency Domain Gating Filtering Method for Hologram

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
  • Fresnel hologram frequency domain gating filtering method based on deep learning
  • Fresnel hologram frequency domain gating filtering method based on deep learning
  • Fresnel hologram frequency domain gating filtering method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Such as figure 1 As shown, a Fresnel hologram frequency-domain gating filtering method based on deep learning is characterized in that:

[0028] (1) Off-axis Fresnel digital holograms of images generated by simulation (such as figure 2 shown), the spectrogram generated by the off-axis Fresnel digital hologram after Fourier transform (FFT) is used as the sample required for network training, and the spectrogram dataset is labeled.

[0029] In this embodiment, the data set is a mixture of 1500 multi-angle reference light Mnist handwritten data set hologram spectrograms and 500 actual pictures from the Internet, of which 1500 marked negative level handwritten data set spectrograms are used as training set, the remaining 200 as a test set, and 300 as a validation set. Assume that the light wavelength of the light source is λ=632.8nm, the distance from the object to the holographic recording surface is 0.3086m, the size of the assigned diffraction surface is 5mm, and ther...

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 Fresnel hologram frequency domain gating filtering method based on deep learning, and the method comprises the steps: employing a spectrogram generated after the Fourier transform of an off-axis Fresnel digital hologram generated through simulation as a sample needed by network training; enabling the feature extraction network to realize extraction of main reconstruction information of the holographic image by learning a relation between zero level and positive and negative level of holographic spectrograms at different angles; finally, filtering the part containing the effective information, performing Fourier inversion to obtain an interference-free hologram, performing hologram reconstruction to obtain a hologram only containing a negative level, and performing compression quality evaluation on analog data. According to the method, the hologram without zero order and other interferences can be generated, the hologram compression rate is improved, and the problems that real-time reconstruction of the hologram is difficult, the reconstruction effect is poor and the compression rate is insufficient are solved. The method has a wide application prospect in the fields of holographic three-dimensional display, holographic projection, information security, digital microscopic holography, intelligent transportation, medical treatment and the like.

Description

technical field [0001] The invention relates to a Fresnel hologram frequency-domain gating filtering method based on deep learning, which is applicable to fields such as holographic three-dimensional display, holographic projection, information security, digital micro-holography, intelligent transportation, and medical treatment. Background technique [0002] With the development of optical technology and the advent of the information age, holography has been widely used in different fields such as real-time three-dimensional display, holographic projection, information security, and digital micro-holography due to its high precision and fast imaging speed. In recent years, digital holography has been widely used in three-dimensional display technology, resulting in an exponential growth in the amount of data. The huge amount of data and complex calculations make real-time transmission and processing very difficult. How to save the storage space of holograms and reduce the t...

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): G06T17/00G06T5/00G06N3/04G06N3/08
CPCG06T17/00G06N3/08G06T2207/20056G06T2207/30168G06T2207/20081G06N3/045G06T5/70
Inventor 桂进斌吴佳雪
Owner KUNMING 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