Deep learning assisted OAM modes demultiplexing

Raskatla, Venugopal; Kumar, Vijay

doi:10.1117/12.2615170

Cited by 7 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we are presenting speckle-based models employing convolutional neural network (CNN) for the recognition of structured light modes [6][7][8][9] . These speckle-learned models overcome the limitation of previous machine-learning techniques by further lifting the alignment constraint.…”

Section: Introductionmentioning

confidence: 99%

Machine learning meets singular optics: speckle-based structured light demultiplexing

Raskatla,

Badavath,

Kumar

2024

Sixteenth International Conference on Correlation Optics

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Machine learning meets singular optics: speckle-based structured light demultiplexing

Raskatla,

Badavath,

Kumar

2024

Sixteenth International Conference on Correlation Optics

View full text Add to dashboard Cite

“…Advancing our technique on speckle-learned OAM recognition [20][21][22][23][24][25], to best of our knowledge, for the first time, here we are presenting a comprehensive analysis of the supervised learning classification of OAM beams using nanostructures focusing on their evolution from near-field (Fresnel region) to far-field (Fraunhofer region). Our approach involves generating speckles of OAM beams by interacting them with microstructures and nanostructures, which exhibit distinctly different characteristics in the near-field and far-field [26].…”

Section: Introductionmentioning

confidence: 99%

Machine-learning-assisted orbital angular momentum recognition using nanostructures

Sharma,

Badavath,

Supraja

et al. 2024

J. Opt. Soc. Am. A

View full text Add to dashboard Cite

The recognition of orbital angular momentum (OAM) in light beams holds significant importance in optical communication. The majority of current OAM recognition techniques are highly sensitive to stringent alignment issues. The speckle-based OAM recognition method reported in [J. Opt. Soc. Am. A 39, 759 (2022)] is alignment-free in the transverse direction of light propagation and has been shown to operate successfully in the far-field region using macrostructures. This study introduces a proof-of-concept for speckle-learned OAM recognition with nanostructures, relaxing the strict alignment requirements in both the transverse and along the direction of light propagation. When the OAM beam interacts with random inhomogeneities at micron and/or nanoscale, it generates an OAM speckle field. Initially, a comprehensive examination of the dynamic evolution of OAM speckle fields, ranging from near field to far field, has been conducted using a ground glass diffuser, featuring random phase inhomogeneities at the micron scale. Subsequently, the investigation proceeds to randomly grown ZnO nanosheets on an aluminum substrate. To achieve rapid and precise OAM recognition, a tailored 3-layer CNN is trained and tested on OAM speckle fields ranging from near-field to far-field to attain an accuracy surpassing 92% . This research expands the technique's applicability, enabling recognition of OAM across near-field to far-field regimes, while leveraging micro-to nanostructures.

show abstract

“… – 21 Despite better accuracy, these are alignment limited and one needs to capture the entire mode. These limitations were overcome in the recent demonstration on the speckle-based CNN 22 – 24 and wavelet scattering network 25 …”

Section: Introductionmentioning

confidence: 99%

“…The LGp,l modes (with p=0) are the special subset of OAM beams described by radial index p and azimuthal index l (also called a charge). The LGp,l beams of the order p and ±l have identical intensity distributions therefore they cannot be distinguished by their sole intensity images or their respective two-dimensional (2D) far-field speckle patterns 22 – 25 as well. A lot of work has been developed to measure and quantify the charge of these modes.…”

Section: Introductionmentioning

confidence: 99%

“…[14][15][16][17][18][19][20][21] Despite better accuracy, these are alignment limited and one needs to capture the entire mode. These limitations were overcome in the recent demonstration on the speckle-based CNN [22][23][24] and wavelet scattering network. 25 The LG p;l modes (with p ¼ 0) are the special subset of OAM beams described by radial index p and azimuthal index l (also called a charge).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Speckle-learned convolutional neural network for the recognition of intensity degenerate orbital angular momentum modes

2023

Self Cite

View full text Add to dashboard Cite

.Intensity degenerate orbital angular momentum (OAM) modes are impossible to recognize by direct visual inspection even using available machine learning techniques. We are reporting speckle-learned convolutional neural network (CNN) for the recognition of intensity degenerate Laguerre–Gaussian (LGp , l) modes, intensity degenerate LG superposition modes, and intensity degenerate perfect optical vortices. The CNN is trained on the simulated one-dimensional far-field intensity speckle patterns of the corresponding intensity degenerate OAM modes. The trained CNN recognizes intensity degenerate OAM modes with an accuracy >99 % . Speckle-learned CNNs are also capable of recognizing intensity degenerate OAM modes even under the presence of high Gaussian white noise and atmospheric turbulence with an accuracy >97 % .

show abstract

Deep learning assisted OAM modes demultiplexing

Cited by 7 publications

References 8 publications

Machine learning meets singular optics: speckle-based structured light demultiplexing

Machine learning meets singular optics: speckle-based structured light demultiplexing

Machine-learning-assisted orbital angular momentum recognition using nanostructures

Speckle-learned convolutional neural network for the recognition of intensity degenerate orbital angular momentum modes

Contact Info

Product

Resources

About