Robust light beam diffractive shaping based on a kind of compact all-optical neural network

Dong, Wei; Hu, Cong; Chen, Mingce; Liu, Kewei; Luo, Jun; Zhang, Xinyu

doi:10.1364/oe.419123

Cited by 34 publications

(19 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To create an all-optical QPI solution without any digital phase reconstruction algorithm, we designed diffractive networks [54][55][56][57][58] that transform the phase information of the input sample into an output intensity pattern, quantitatively revealing the object phase distribution through an intensity recording. Figure 1 illustrates the schematic of a 5-layer diffractive network that was trained to all-optically synthesize the QPI signal of a given input phase object (see the Experimental Section for training details).…”

Section: Resultsmentioning

confidence: 99%

All‐Optical Phase Recovery: Diffractive Computing for Quantitative Phase Imaging

Mengü

Ozcan

2022

Advanced Optical Materials

View full text Add to dashboard Cite

Quantitative phase imaging (QPI) is a label‐free computational imaging technique that provides optical path length information of specimens. In modern implementations, the quantitative phase image of an object is reconstructed digitally through numerical methods running in a computer, often using iterative algorithms. Here, a diffractive QPI network that can perform all‐optical phase recovery is demonstrated, and the quantitative phase image of an object is synthesized by converting the input phase information of a scene into intensity variations at the output plane. A diffractive QPI network is a specialized all‐optical processor designed to perform a quantitative phase‐to‐intensity transformation through passive diffractive surfaces that are spatially engineered using deep learning and image data. Forming a compact, all‐optical network that axially extends only ≈200–300λ, where λ is the illumination wavelength, this framework can replace traditional QPI systems and related digital computational burden with a set of passive transmissive layers. All‐optical diffractive QPI networks can potentially enable power‐efficient, high frame‐rate, and compact phase imaging systems that might be useful for various applications, including, e.g., microscopy and sensing.

show abstract

Section: Resultsmentioning

confidence: 99%

All‐Optical Phase Recovery: Diffractive Computing for Quantitative Phase Imaging

Mengü

Ozcan

2022

Advanced Optical Materials

View full text Add to dashboard Cite

show abstract

“…The idea behind this technology is to interpret a sequence of optical elements such as DOEs and LCoS as the physical realization of an artificial neural network in which each optic represents a layer and the layers are interconnected by wave-optical propagation. While diffractive neural networks have mainly been used for image classification [21] and only few publications treat beam shaping [22], we exploit the full flexibility of the high number of degrees of freedom for beam shaping. The design approach is to train the diffractive neural network on a computer using AI approaches and afterwards realize the respective optical elements.…”

Section: Diffractive Neural Networkmentioning

confidence: 99%

“…Due to the design by a training approach, multiple requirements in addition to a prescribed intensity distribution can be included. Examples are a robustness regarding misalignment [22], additional phase optimization, and multi target/intensity optimization. Figure 4 shows the simulation results for a diffractive neural network which is designed to not only optimize the intensity in the target plane but also flatten the electromagnet field's phase.…”

Section: Diffractive Neural Networkmentioning

confidence: 99%

Application-adapted beam shaping with cutting-edge optical elements in laser materials processing

Völl

Buske

Stollenwerk

et al. 2023

High-Power Laser Materials Processing: Applications, Diagnostics, and Systems XII

View full text Add to dashboard Cite

The quality, efficiency and robustness of laser materials processing significantly benefits from an application-adapted beam shaping. For a successful realization of tailored beam shapes that utilizes state-of-the-art optic concepts, research in two different steps is necessary. First, information about the required beam shape (especially in terms of intensity distribution) is needed. During the last years, we have developed and validated a method to solve the so-called inverse heat conduction problem which enables the derivation of an intensity distribution that specifically tailors the induced temperature profile within the processed material. Here, we present the latest enhancement of the algorithm to complex time-dependent scenarios and new applications, e.g. from the field of surface treatment and tape placement. With the knowledge of the target beam shape, in a second step, optical systems must be designed that enable the realization of the achieved, highly complex intensity distributions. We further demonstrate the potential and limits of novel optics such as freeform mirrors, LCoS-SLMs or diffractive neural networks as well as VCSELs for application-adapted beam shaping. We especially present specific design methods that can contribute to a robust, flexible, and practical realization in various applications.

show abstract

“…These advantages make optical neural networks an attractive solution for applications that require fast and efficient information processing, such as real-time image and video processing, autonomous systems, and communication networks. [8][9][10][11][12][13][14][15] Optical neural networks were first reported in the 1980s but have gained popularity in recent years due to advancements in technology. Optical neural networks can be implemented as either free space [45][46][47][48][49][50][51][52][53][54][55] or integrated [56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75] versions, each with their own advantages and disadvantages.…”

Section: Introductionmentioning

confidence: 99%

Black-box simulation method: train the optical model from output

Hu¹,

Nouri²,

Dalir³

et al. 2023

Physics and Simulation of Optoelectronic Devices XXXI

View full text Add to dashboard Cite

Accurate simulation is a critical requirement for modern optical systems, but precise theoretical modeling can be challenging due to various factors such as misalignment, theoretical approximations, and instrument errors. This paper presents the black-box simulation method, which addresses these limitations by training the optical system model using the optical field outputs. This approach leads to more accurate simulations, making it possible to effectively train optical Neural Network systems for improved performance.

show abstract

Robust light beam diffractive shaping based on a kind of compact all-optical neural network

Cited by 34 publications

References 31 publications

All‐Optical Phase Recovery: Diffractive Computing for Quantitative Phase Imaging

All‐Optical Phase Recovery: Diffractive Computing for Quantitative Phase Imaging

Application-adapted beam shaping with cutting-edge optical elements in laser materials processing

Black-box simulation method: train the optical model from output

Contact Info

Product

Resources

About