Machine learning based adaptive optics for doughnut-shaped beam

Zhang, Yiye; Wu, Chenxue; Song, Yanchun; Si, Ke; Zheng, Yao; Hu, Ling; Chen, Jiajia; Tang, Liang; Gong, Wei

doi:10.1364/oe.27.016871

Cited by 24 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML techniques have been used to create self-tuning, mode-locked lasers; [4][5][6] for automating the production of Bose-Einstein condensation; 7 and maintaining doughnut-shaped beams in scattering media. 8 ML has recently even been used to create new quantum experiments: the system both learned to create a variety of entangled states and improved the efficiency of their realization. 9 Despite these advances, no work using ML for beam alignment has been found.…”

Section: Introductionmentioning

confidence: 99%

The Raspberry Pi auto-aligner: Machine learning for automated alignment of laser beams

Mathew

O’Donnell

Pizzey

et al. 2021

Review of Scientific Instruments

View full text Add to dashboard Cite

We present a novel solution to automated beam alignment optimization. This device is based on a Raspberry Pi computer, stepper motors, commercial optomechanics and electronic devices, and the open-source machine learning algorithm M-LOOP. We provide schematic drawings for the custom hardware necessary to operate the device and discuss diagnostic techniques to determine the performance. The beam auto-aligning device has been used to improve the alignment of a laser beam into a single-mode optical fiber from manually optimized fiber alignment, with an iteration time of typically 20 minutes. We present example data of one such measurement to illustrate device performance.

show abstract

Section: Introductionmentioning

confidence: 99%

The Raspberry Pi auto-aligner: Machine learning for automated alignment of laser beams

Mathew

O’Donnell

Pizzey

et al. 2021

Review of Scientific Instruments

View full text Add to dashboard Cite

show abstract

“…Hu et al presented a learning-based Shack-Hartmann wavefront sensor to detect the highorder aberrations without image segmentation or centroid positioning [20]. What is more, Zhang et al demonstrated a machine learningbased AO method to recover the doughnut-shaped beam with numerous optical control elements and fast operation speed [21].…”

Section: Introductionmentioning

confidence: 99%

Adaptive optics for structured illumination microscopy based on deep learning

Zheng

Chen

et al. 2021

Cytometry Pt A

Self Cite

View full text Add to dashboard Cite

Structured illumination microscopy (SIM) is widely used in biological imaging for its high resolution, fast imaging speed, and simple optical setup. However, when imaging thick samples, the structured illumination patterns in SIM will suffer from optical aberrations, leading to a serious deterioration in resolution. Therefore, it is necessary to reconstruct structured illumination patterns with high quality and efficiency in deep tissue imaging. Here we demonstrate an adaptive optics (AO) correction method based on deep learning in wide-field SIM imaging system. The mapping between the coefficients of the first 15 Zernike modes and their corresponding distorted patterns is established to train the convolution neural network (CNN). The results show that the optimized CNN can predict the aberration phase within 10.1 ms with a personal computer. The correlation index between the aberration phases and their corresponding predicted aberration phase is up to 0.9986. This method is highly robust and effective for patterns with various spatial densities and illumination conditions and able to effectively correct the imaging distortion caused by optical aberration in SIM system.

show abstract

“…Aside from the typical Gaussian intensity profile and the highly accurate top hat profile, non-conventional laser beam shapes are ever more required for specific applications [35]. Annular beams, in particular the doughnut-shaped profile (TEM 01 * mode), may broaden the laser technology applications, especially at nanoscale, extending laser materials processing techniques, improving lithography accuracy, and creating novel structures in materials [11,36].…”

Section: Introductionmentioning

confidence: 99%

Doughnut-Shaped and Top Hat Solar Laser Beams Numerical Analysis

et al. 2021

View full text Add to dashboard Cite

Aside from the industry-standard Gaussian intensity profile, top hat and non-conventional laser beam shapes, such as doughnut-shaped profile, are ever more required. The top hat laser beam profile is well-known for uniformly irradiating the target material, significantly reducing the heat-affected zones, typical of Gaussian laser irradiation, whereas the doughnut-shaped laser beam has attracted much interest for its use in trapping particles at the nanoscale and improving mechanical performance during laser-based 3D metal printing. Solar-pumped lasers can be a cost-effective and more sustainable alternative to accomplish these useful laser beam distributions. The sunlight was collected and concentrated by six primary Fresnel lenses, six folding mirror collectors, further compressed with six secondary fused silica concentrators, and symmetrically distributed by six twisted light guides around a 5.5 mm diameter, 35 mm length Nd:YAG rod inside a cylindrical cavity. A top hat laser beam profile (Mx2 = 1.25, My2 = 1.00) was computed through both ZEMAX® and LASCAD® analysis, with 9.4 W/m2 TEM00 mode laser power collection and 0.99% solar-to-TEM00 mode power conversion efficiencies. By using a 5.8 mm laser rod diameter, a doughnut-shaped solar laser beam profile (Mx2 = 1.90, My2 = 1.00) was observed. The 9.8 W/m2 TEM00 mode laser power collection and 1.03% solar-to-TEM00 mode power conversion efficiencies were also attained, corresponding to an increase of 2.2 and 1.9 times, respectively, compared to the state-of-the-art experimental records. As far as we know, the first numerical simulation of doughnut-shaped and top hat solar laser beam profiles is reported here, significantly contributing to the understanding of the formation of such beam profiles.

show abstract

Machine learning based adaptive optics for doughnut-shaped beam

Cited by 24 publications

References 28 publications

The Raspberry Pi auto-aligner: Machine learning for automated alignment of laser beams

The Raspberry Pi auto-aligner: Machine learning for automated alignment of laser beams

Adaptive optics for structured illumination microscopy based on deep learning

Doughnut-Shaped and Top Hat Solar Laser Beams Numerical Analysis

Contact Info

Product

Resources

About