Ion beam figuring high gradient optical aspherical surface

Tang, Wa; Deng, Wei; Yin, Xiaolin

doi:10.1117/12.2504901

Cited by 3 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In response, researchers have embraced the potential of lightwave technology, harnessing its attributes of remarkable speed, wide bandwidth, and minimal crosstalk. Optical neural networks [1][2][3][4][5][6][7], encompassing systems based on silicon photonics and diffractive surfaces, have emerged as promising avenues for achieving rapid object detection, as well as facilitating efficient backpropagation computation and beam steering.…”

Section: Introductionmentioning

confidence: 99%

Robust diffractive neural networks for resilient all-optical intelligent imaging detection

shi,

zhang

2024

MIPPR 2023: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications

View full text Add to dashboard Cite

Conventional imaging detection relies on electrical modulation, but it grapples with limitations in speed and adaptability within complex environments. Intelligent imaging detection, capable of manipulating micro-nano-lightfields generated by imaging objective lenses, is essential. The all-optical deep diffractive neural network (D2NN) offers a potential solution, enabling swift spatial information transformation and recognition. However, existing D2NN versions are vulnerable to optical disturbances due to their reliance on computer-based backpropagation. This paper introduces a robust diffractive neural network (SRNN) mathematical model designed to mitigate the impact of structural errors and lightwave frequency shifting without necessitating hardware modifications. The SRNN model incorporates Gaussian noise at various levels during training to emulate the error distribution of the phase mask, augmenting the network's resilience to hardware errors. Weight-Noise-Injection training assists the network in locating an error-insensitive region, thereby enhancing its robustness.

show abstract

Section: Introductionmentioning

confidence: 99%

Robust diffractive neural networks for resilient all-optical intelligent imaging detection

shi,

zhang

2024

MIPPR 2023: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications

View full text Add to dashboard Cite

show abstract

“…In order to address the challenges posed by complex imaging tasks, we present a pioneering paradigm that combines the principles of diffractive networks [1][2][3][4][5] for encoding and electronic networks for decoding. This novel framework, referred to as DE-ED, establishes a learning-based computational imaging approach that attains the better results in advanced imaging applications that involve high dynamic range and all-focus.…”

Section: Introductionmentioning

confidence: 99%

Enhancing All-in-Focus Imaging with Opto-Electronic Computational Camera using DE-ED Paradigm

Shi,

Zhou,

Zhang

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

A novel approach called DE-ED (Diffractive Encoding-Electronic Decoding) for rapid all-in-focus capturing is presented in this paper. The paradigm combines the strengths of diffractive networks (DON) for encoding and electronic-based networks (ENN) for decoding, resulting in a robust computational imaging framework driven by advanced deep learning techniques. By leveraging the collaboration between diffractive layers and electronic layers, our model achieves remarkable enhancements in the ability to generalize and fit data. This means that the model can effectively capture and process complex visual information, resulting in superior image quality and enhanced performance.

show abstract