Machine learning technologies have been extensively applied in high-performance information-processing fields. However, the computation rate of existing hardware is severely circumscribed by conventional Von Neumann architecture. Photonic approaches demonstrated extraordinary potential for executing deep processes that involve complex calculations. In this work, an on-chip diffractive optical neural network (DONN) based on a silicon-on-insulator platform proposed to perform machine tasks...

Machine vision faces bottlenecks in computing power consumption and large amounts of data. Although opto-electronic hybrid neural networks can provide assistance, they usually have complex structures are highly dependent on a coherent light source; therefore, not suitable for natural lighting environment applications. In this paper, we propose novel lensless network architecture machine The optimizes passive optical mask by means task-oriented design, performs the convolution calculation...

An integrated physical diffractive optical neural network (DONN) is proposed based on a standard silicon-on-insulator (SOI) substrate. This DONN has compact structure and can realize the function of machine learning with whole-passive fully-optical manners. The designed by spatial domain electromagnetic propagation model, approximate process neuron value mapping optimized well to guarantee consistence between pre-trained SOI integration implementation. model better ensure manufacturability...

Ever-growing deep-learning technologies are making revolutionary changes for modern life. However, conventional computing architectures designed to process sequential and digital programs but burdened with performing massive parallel adaptive applications. Photonic integrated circuits provide an efficient approach mitigate bandwidth limitations the power-wall brought on by its electronic counterparts, showing great potential in ultrafast energy-free high-performance computation. Here, we...

Livestock-associated methicillin-resistant Staphylococcus aureus (LA-MRSA) is an important zoonotic microorganism that increasingly causing public health concern worldwide. The objective of this study was to determine the transmission and occurrence MRSA in a slaughterhouse environment evaluate its antimicrobial resistance genetic characterization. In study, we conducted comprehensive epidemiological survey S. by spa typing whole-genome sequencing (WGS) samples obtained from pork production...

An opto-electronic neural network is designed for video object detection from a long-exposure blurred image. This combines an optical encoder, convolutional decoder, and module, which are jointly optimized end-to-end. The joint loss adopted updating the according to physical constraints of hardware via back-propagation. A high-speed refreshed spatial light modulator used as encoder part generate coded sub-images, then, single image obtained after common camera. rest detection. Both...

Snapshot compressive imaging (SCI) encodes high-speed scene video into a snapshot measurement and then computationally makes reconstructions, allowing for efficient high-dimensional data acquisition. Numerous algorithms, ranging from regularization-based optimization deep learning, are being investigated to improve reconstruction quality, but they still limited by the ill-posed information-deficient nature of standard SCI paradigm. To overcome these drawbacks, we propose new key frames...

Non-line-of-sight (NLOS) imaging is an emerging technique for detecting objects behind obstacles or around corners. Recent studies on passive NLOS mainly focus steady-state measurement and reconstruction methods, which show limitations in recognition of moving targets. To the best our knowledge, we propose a novel event-based method. We acquire asynchronous data contains detailed dynamic information target, efficiently ease degradation speckle caused by movement. Besides, create first...

Staphylococcus aureus, especially methicillin-resistant S. aureus (MRSA), is an important zoonotic microorganism that increasingly causes public health concern worldwide. The objective of this study was to determine the prevalence and transmission in duck farms evaluate its antimicrobial resistance genetic characteristics.

Integrated diffractive optical neural networks (DONNs) have significant potential for complex machine learning tasks with high speed and ultralow energy consumption. However, the on-chip implementation of a high-performance network is limited by input dimensions. In contrast to existing photonic networks, space-time interleaving technology based on arrayed waveguides designed realize an DONN high-speed, high-dimensional, all-optical signal modulation. To demonstrate performance high-speed...

Abstract Machine learning technologies have been extensively applied in high-performance information-processing fields. However, the computation rate of current hardware is severely circumscribed by conventional Von Neumann architecture. Photonic approaches demonstrated extraordinary potential for executing deep involving complex calculations. In this work, an on-chip diffractive optical neural network (DONN) based on a silicon-on-insulator (SOI) platform proposed to perform machine tasks...

For moving objects, 3D mapping and tracking has found important applications in the reconstruction for vision odometry or simultaneous localization mapping. This paper presents a novel camera architecture to locate fast-moving objects four-dimensional (4D) space ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" id="m1"> <mml:mrow> <mml:mi>x</mml:mi> </mml:mrow> </mml:math> , id="m2"> <mml:mi>y</mml:mi> id="m3"> <mml:mi>z</mml:mi> id="m4"> <mml:mi>t</mml:mi> )...

Abstract Background The adverse effects of cardiopulmonary bypass during open cardiac surgery, including hemodilution, seem to be inevitable, especially for patients who generally have a relatively lower BMI with small blood volumes. This study reports the modification and use (CPB) system reduce priming volume hemodilution. Methods is retrospective 462 adult underwent valve replacement surgery from January 2019 September 2021 at General Hospital Western Theater Command. modified group...

A miniature on-chip diffractive optical neural network (DONN) with a footprint of 1 . 35 × 10 – 3 mm 2 is designed through the particle swarm optimization (PSO) algorithm, yielding an accuracy 93.3% in simulation on Iris plants dataset.

The novel camera architecture facilitates the development of machine vision. Instead capturing frame sequences in temporal domain as traditional video cameras, FourierCam directly measures pixel-wise spectrum a single shot through optical coding. Compared to classic cameras and time-frequency transformation pipeline, this programmable frequency-domain sampling strategy has an attractive combination characteristics for low detection bandwidth, computational burden, data volume. Based on...

The framework provides a novel pipeline for action recognition. recognition task classifies the label of scene. High-speed cameras are commonly used to generate high frame-rate videos capturing sufficient motion information. However, data volume would be bottleneck system. With insight that discrete cosine transform (DCT) video signals reveals information remarkably, instead obtaining as with traditional cameras, proposed method directly captures DCT spectrum in single shot through optical...

Ghost imaging is a fascinating framework which constructs the image of an object by correlating measurements between received beams and reference beams, none carries structure information independently. Recently, taking into account spacetime duality in optics, computational temporal ghost has attracted attention. Here, we propose novel Fourier (FTGI) scheme to achieve single-shot nonreproducible signals. By sinusoidal coded modulation, images are obtained recovered applying transformation....

Objectives: Cardiopulmonary bypass (CPB) induces inflammatory homeostasis dysregulation, closely related to many postoperative adverse effects. Minimizing the systemic response CPB is imperative improving cardiac surgery safety. This study aimed retrospectively evaluate efficacy of hemoperfusion cartridge, a device recently designed for extracorporeal blood purification remove cytokines from patients undergoing valve replacement using CPB. Methods: The (HP) group consisted 138 patients, who...

With the growing popularity of smartphone photography in recent years, web photos play an increasingly important role all walks life. Source camera identification aims to establish a reliable linkage from captured images their source cameras, and has broad range applications, such as image copyright protection, user authentication, investigated evidence verification, etc. This paper presents innovative practical framework that employs neural-network enhanced sensor pattern noise trace back...