Holographic coding method based on neural network

A technology of neural network and coding method, which is applied in the direction of neural learning method, biological neural network model, neural architecture, etc., can solve problems such as the inability to realize complex amplitude light field reconstruction, lack of complex amplitude modulation devices, and low quality holograms, and achieve Effects of improving computing speed, reducing computing times and data storage capacity, and improving computing efficiency

Pending Publication Date: 2021-11-16
SOUTHEAST UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Due to the lack of complex amplitude modulation devices capable of simultaneously modulating amplitude and phase, the currently widely used algorithms in the field of digital holography can only control the light field information in one dimension, while the other dimension is random and uncontrollable, resulting in the difficulty of reconstructing holograms. low quality
The lack of complex amplitude modulation devices makes it impossible to realize real complex amplitude light field reconstruction, which makes the current holographic display technology face a huge bottleneck

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
  • Holographic coding method based on neural network
  • Holographic coding method based on neural network
  • Holographic coding method based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0098] The target complex amplitude is decomposed into two channel information according to the real part and the imaginary part, which is used as the input of the neural network. And use the same method to process the training data set to train the ResNet network structure, and the target output is the phase holographic encoding result. After the phase-type holographic encoding result is obtained through network calculation, it is loaded onto the phase-type spatial light modulator Holoeye Pluto produced by Holoeye in Germany through the computer. The specification of this phase-type spatial light modulator is 1920×1080 pixels, and the pixel pitch is 8 μm. The light source is monochromatic green light with a wavelength of 532nm emitted by a monochromatic laser. After the light emitted by the spatial light modulator passes through the dichroic prism, it passes through a filter system composed of two lenses and a diaphragm, such as Figure 4 , f 1 =f 2 =10cm, the size of the...

Embodiment 2

[0101]The target complex amplitude is decomposed into two channel information according to the amplitude distribution and phase distribution, which is used as the input of the neural network. And use the same method to process the training data set to train the ResNet network structure, and the target output is the phase holographic encoding result. After the phase-type holographic encoding result is obtained through network calculation, it is loaded onto the phase-type spatial light modulator Holoeye Pluto produced by Holoeye in Germany through the computer. The specification of this phase-type spatial light modulator is 1920×1080 pixels, and the pixel pitch is 8 μm. The light source is monochromatic green light with a wavelength of 532nm emitted by a monochromatic laser. After the light emitted by the spatial light modulator passes through the dichroic prism, it passes through a filter system composed of two lenses and a diaphragm, such as Figure 4 , f 1 =f 2 =10cm, the...

Embodiment 3

[0104] The target complex amplitude is decomposed into two complex amplitude distributions with the same amplitude distribution and different phase distribution, which are used as the input of the neural network. And use the same method to process the training data set to train the ResNet network structure, and the target output is the phase holographic encoding result. After the phase-type holographic encoding result is obtained through network calculation, it is loaded onto the phase-type spatial light modulator Holoeye Pluto produced by Holoeye in Germany through the computer. The specification of this phase-type spatial light modulator is 1920×1080 pixels, and the pixel pitch is 8 μm. The light source is monochromatic green light with a wavelength of 532nm emitted by a monochromatic laser. After the light emitted by the spatial light modulator passes through the dichroic prism, it passes through a filter system composed of two lenses and a diaphragm, such as Figure 4 , ...

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 holographic coding method based on a neural network, and the method comprises the steps: decomposing a target complex amplitude into real data of two channels, and using the real data as the input of a neural network model; processing the real data through the neural network model to obtain a single-channel holographic coding result; loading the obtained single-channel holographic coding result to a spatial light modulator to obtain output light; and constructing an optical system, propagating the obtained output light through the optical system, and finally obtaining a target complex amplitude result recovered by the single-channel holographic coding graph. Holographic reconstruction is carried out on the generated holographic coding result, the reconstructed holographic result has smaller noise, meanwhile, the method can be widely applied to holographic coding and holographic reconstruction in a complex amplitude scene, and a new thought is provided for further development of holographic display.

Description

technical field [0001] The invention relates to hologram encoding, in particular to a neural network-based holographic encoding method. Background technique [0002] Holographic display technology is a technology that can reconstruct the complex amplitude light field of a three-dimensional scene, and is therefore considered to be the most ideal next-generation stereoscopic display technology. With the continuous development of computer technology and liquid crystal display technology, as long as a three-dimensional scene can be described mathematically, we can get rid of the complicated interference recording process in traditional holography, without using the recording dry plate of the high-coherence light source system, and instead Holograms are numerically computed in a computer to reconstruct real or post-synthesized scenes. [0003] Due to the lack of complex amplitude modulation devices capable of simultaneously modulating amplitude and phase, currently widely used a...

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): G06T17/00G06N3/04G06N3/08G06T15/50
CPCG06T17/00G06T15/50G06N3/084G06N3/045
Inventor 王翘楚蔡玉彤夏军
Owner SOUTHEAST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap