Search

Frequency blue shift of optical solitons, due to light-plasma interactions is an important nonlinear process of laser frequency up-conversion in gas-filled hollow-core fibers, which can be applied for generating ultrafast pulses with tunable wavelengths covering infrared and visible regimes. While the mechanism of photoionization in this frequency up-conversion process has been well studied for a few years, the spectral width tunability of output ultrafast pulses generated in such a hollow-core fiber system has not been comprehensively investigated yet. Here, we demonstrate that the theory of adiabatic soliton propagation could be used to understand the spectral bandwidth evolution of photoionization-induced blue-shifted solitons in gas-filled hollow-core anti-resonant fibers. Experimentally, through varying the gas pressure in the hollow-core fiber and the input pulse energy, high-repetition-rate (10 kHz) tunable ultrafast pulses were obtained at the fiber output with a central wavelength tuning range from ∼900 nm to ∼650 nm and a bandwidth tuning range (at 700 nm) from ∼100 nm to ∼180 nm, corresponding to chirp-compensated pulse width tuning from 7.2 fs to 5.4 fs. These experimental results agree well with the theoretical predictions, and the femtosecond visible light source with high-flexible pulse parameters, demonstrated here, may have some potentials in ultrafast spectroscopy and nonlinear-optics applications.

Time-stretched dispersive Fourier transform (TS-DFT) technique has been widely used for studying transient soliton dynamics due to its unique capability of direct time-frequency mapping of ultrafast pulses in the real time. Particularly, multi-soliton structures that are ubiquitous in ultrafast lasers have been routinely analyzed using TS-DFT technique, giving concurrently spacing and relative-phase information during soliton interactions. While the TS-DFT signal recorded by the fast electronics would inevitably deviate from the genuine spectral interferogram, the consequent discrepancies have not, however, been systematically interrogated yet. In this work, we performed detailed comparisons between the temporal TS-DFT signal of multi-soliton structures (with tunable spacings and phases) and their genuine spectral interferograms measured using optical spectrum analyzer. We revealed that due the impulsive response of the detection system, the retrieved relative-phases from the TS-DFT signal could have non-trivial discrepancies from their actual values, even though the bandwidth of the detection electronics is a few times broader than that of the TS-DFT signal. Our findings can help to provide clearer insights in multi-soliton dynamics using TS-DFT technique.

In many applications related to point clouds, registration is an inevitable step when processing point cloud data. The registration methods performed by local shape descriptor (LSD) are computationally efficient and suitable for different scenes, but they achieve low registration accuracy due to the limited performance of the LSD. For this reason, a novel LSD is designed for terrestrial point clouds. First, a simple yet efficient local reference frame (LRF) is developed. The LRF is calculated by the robust normal vector and constant vector, so it has high repeatability. This increases the robustness of the descriptor. Then, the local neighborhood information is encoded based on the LRF in 3-D space. The voxel centers are used to compute the feature descriptor. This increases the descriptiveness of the descriptor because the voxel centers can well preserve the local information. Thus, the proposed LSD is highly descriptive and strongly robust. Based on the novel LSD, a registration method is given. The proposed LSD makes the registration method have high accuracy. Also, the proposed LRF can improve the performance of the correspondence selection, which is an important process in an LSD-based registration method. The experiments performed on the point clouds of different scenes well illustrate that our LRF method has significantly better repeatability and robustness in comparison with other LRF methods. The proposed LRF can largely improve the performance of the descriptor. Our LSD also has significantly better descriptiveness and robustness compared to the other descriptors. As a result, our registration method achieves high accuracy and good time efficiency due to the proposed LSD. The code will be available at https://github.com/taowuyong?tab=repositories after publication.

Parkinson's disease (PD) is a common degenerative disease of the nervous system in the elderly. The early diagnosis of PD is very important for potential patients to receive prompt treatment and avoid the aggravation of the disease. Recent studies have found that PD patients always suffer from emotional expression disorder, thus forming the characteristics of “masked faces”. Based on this, we thus propose an auto PD diagnosis method based on mixed emotional facial expressions in the paper. Specifically, the proposed method is cast into four steps: Firstly, we synthesize virtual face images containing six basic expressions (i.e., anger, disgust, fear, happiness, sadness, and surprise) via generative adversarial learning, in order to approximate the premorbid expressions of PD patients; Secondly, we design an effective screening scheme to assess the quality of the above synthesized facial expression images and then shortlist the high-quality ones; Thirdly, we train a deep feature extractor accompanied with a facial expression classifier based on the mixture of the original facial expression images of the PD patients, the high-quality synthesized facial expression images of PD patients, and the normal facial expression images from other public face datasets; Finally, with the well-trained deep feature extractor, we thus adopt it to extract the latent expression features for six facial expression images of a potential PD patient to conduct PD/non-PD prediction. To show real-world impacts, we also collected a new facial expression dataset of PD patients in collaboration with a hospital. Extensive experiments are conducted to validate the effectiveness of the proposed method for PD diagnosis and facial expression recognition.

Spiking Neural Networks (SNNs) are currently the most widely used computing model for neuroscience communities. There is also an increasing research interest in exploring the potential of SNN in brain-inspired computing, artificial intelligence, and other areas. As SNNs possess distinguished characteristics that originate from biological authenticity, they require dedicated simulation frameworks to achieve usability and efficiency. However, there is no widely-used, easily accessible, high performance SNN simulation framework for GPU clusters. In this paper, we propose ENLARGE, an efficient SNN simulation framework on GPU clusters. ENLARGE provides a multi-level architecture that deals with computation, communication, and synchronization hierarchically. We also propose an efficient communication method with an all-to-all communication pattern. To deal with the delay of spike delivery, which is the most distinguished SNN characteristic, several delay-aware optimization methods are also proposed. We further propose a multilevel workload management method. Various experiments are carried out to demonstrate the performance and scalability of the framework, as well as the effects of the optimization methods. Test results show that ENLARGE can achieve $3.17\times \sim 28.12\times$3.17×∼28.12× speedup compared with the most widely used NEST simulator and $3.26\times \sim 13.57\times$3.26×∼13.57× speedup compared with the widely used NEST GPU simulator for GPU clusters.

Single sample per person face recognition (SSPP FR) is one of the most challenging problems in FR due to the extreme lack of enrolment data. To date, the most popular SSPP FR methods are the generic learning methods, which recognize query face images based on the so-called prototype plus variation (i.e., P+V) model. However, the classic P+V model suffers from two major limitations: 1) it linearly combines the prototype and variation images in the observational pixel-spatial space and cannot generalize to multiple nonlinear variations, e.g., poses, which are common in face images and 2) it would be severely impaired once the enrolment face images are contaminated by nuisance variations. To address the two limitations, it is desirable to disentangle the prototype and variation in a latent feature space and to manipulate the images in a semantic manner. To this end, we propose a novel disentangled prototype plus variation model, dubbed DisP+V, which consists of an encoder–decoder generator and two discriminators. The generator and discriminators play two adversarial games such that the generator nonlinearly encodes the images into a latent semantic space, where the more discriminative prototype feature and the less discriminative variation feature are disentangled. Meanwhile, the prototype and variation features can guide the generator to generate an identity-preserved prototype and the corresponding variation, respectively. Experiments on various real-world face datasets demonstrate the superiority of our DisP+V model over the classic P+V model for SSPP FR. Furthermore, DisP+V demonstrates its unique characteristics in both prototype recovery and face editing/interpolation.

Dispersive waves (DWs) are ubiquitous in nonlinear optical systems that host optical solitons. In particular, DWs in soliton fiber lasers have attracted continuous interest over the past decades since they are closely related to laser soliton instabilities. Meanwhile, intra-cavity DWs also play important roles in solion interactions. While previous studies mostly focused on their spectral properties, characterizations and manipulations of the intra-cavity DWs in the time domain remain unexplored. Here, we report the experimental studies on temporal structures of intra-cavity DWs, unveiling their exponentially-decaying profiles which can be manipulated by varying cavity parameters, following a simple analytic model. We also studied the evolution dynamics of the intra-cavity DWs under abrupt cavity parameter variations. We observed that both the amplitude and the width of the DW waveform went through variations as the pump power was changed, while their evolution trajectories followed intimately that of the soliton energy. These results provide insights into the DW emission mechanism in mode-locked lasers, and the control methods of intra-cavity DWs may find potential applications in the fields of ultrafast pulse generation and soliton dynamics.

Synchronization of mechanical oscillators by optical forces is a topic that has been much explored in recent years, for example, in the context of SiN microdisk resonators. Here we report stable long-term synchronization of the core vibrations of three different photonic crystal fibers, driven intra-cavity by a 2 GHz train of timing-modulated pulses in a high harmonic opto-acoustically mode-locked fiber laser. The core resonances are equally spaced in frequency and are coupled purely by the optical field. Under the correct conditions, they become stably synchronized, being simultaneously driven by the timing-modulated pulse train. Floquet–Bloch theory, in which the pulses are treated as particles trapped in potential wells and coupled by optomechanical back-action, describes the complex temporal dynamics observed in the experiments. This unique system provides a novel means of modifying the temporal structure of pulse trains running at few-gigahertz repetition rates.

This article focuses on a new and practical problem in single-sample per person face recognition (SSPP FR), i.e., SSPP FR with a contaminated biometric enrolment database (SSPP-ce FR), where the SSPP-based enrolment database is contaminated by nuisance facial variations in the wild, such as poor lightings, expression change, and disguises (e.g., wearing sunglasses, hat, and scarf). In SSPP-ce FR, the most popular generic learning methods will suffer serious performance degradation because the prototype plus variation (P+V) model used in these methods is no longer suitable in such scenarios. The reasons are twofold. First, the contaminated enrolment samples could yield bad prototypes to represent the persons. Second, the generated variation dictionary is simply based on the subtraction of the average face from generic samples of the same person and cannot well depict the intrapersonal variations. To address the SSPP-ce FR problem, we propose a novel iterative dynamic generic learning (IDGL) method, where the labeled enrolment database and the unlabeled query set are fed into a dynamic label feedback network for learning. Specifically, IDGL first recovers the prototypes for the contaminated enrolment samples via a semisupervised low-rank representation (SSLRR) framework and learns a representative variation dictionary by extracting the “sample-specific” corruptions from an auxiliary generic set. Then, it puts them into the P+V model to estimate labels for query samples. Subsequently, the estimated labels will be used as feedback to modify the SSLRR, thus updating new prototypes for the next round of P+V-based label estimation. With the dynamic learning network, the accuracy of the estimated labels is improved iteratively by virtue of the steadily enhanced prototypes. Experiments on various benchmark face data sets have demonstrated the superiority of IDGL over state-of-the-art counterparts. [ABSTRACT FROM AUTHOR]

A copper-free Sandmeyer-type fluorosulfonylation reaction is reported. Utilizing Na2S2O5and Selectfluor as the sulfur dioxide and fluorine sources, respectively, aryldiazonium salts were transformed into sulfonyl fluorides. The one-pot direct synthesis of sulfonyl fluorides from aromatic amines was also realized via in situ diazotization. The practicality of this method was demonstrated by the broad functional group tolerance, gram-scale synthesis, and late-stage fluorosulfonylation of natural products and pharmaceuticals.

Here we report a method for the site-selective intermolecular C(sp3)–H amination of carboxamides by merging transition-metal catalysis and the hydrogen atom transfer strategy. The reaction proceeds through a sequence of favorable single-electron transfer, 1,5-hydrogen atom transfer, and C–N cross-coupling steps, thus allowing access to a series of desired products. This reaction could accommodate a wide diversity of nitrogen nucleophiles as well as demonstrate excellent chemoselectivity and functional group compatibility.

Lane changing is one of the basic driving behaviours, which may induce traffic oscillations and incidents. However, it is difficult to well model the lane-changing decision process due to the complex traffic status. To promote the prediction accuracy of lane-changing decisions, this paper presents a probability lane-changing model by taking into account the memory effect. That is, the lane-changing decision model considers a series of trajectory data rather than the data of a specific time utilized in most existing models. Furthermore, the drivers are classified in terms of lane-changing trajectories, which is expected to further promote the prediction accuracy of the lane-changing decision model. Calibrations and validations are carried out based on the NGSIM data, which indicate that the proposed model can significantly promote the prediction accuracy of lane-changing decisions.

Strong enhancement of optoacoustic interactions in the micrometer-sized core of a photonic crystal fiber (PCF) enables stable, harmonic mode locking of a soliton fiber laser at GHz frequencies. Here we report that by tapering the PCF during the draw, the optoacoustic gain bandwidth can be broadened to ∼47  MHz, more than 3 times wider than in the untapered fiber. This made possible broad pulse-repetition-rate tuning over 66 MHz (from 2.042 to 2.108 GHz) of an optoacoustically mode-locked soliton fiber laser. Within this tuning range, the harmonically mode-locked pulse trains at the laser output were observed to be quite robust, with better than 40 dB supermode suppression ratio, sub-ps pulse timing jitter, and <0.2%relative intensity noise. This gigahertz-rate, near-infrared soliton fiber laser has remarkable pulse-rate tunability and low noise level, and has important potential applications in frequency metrology, high-speed optical sampling, and fiber telecommunications.

Buildings in seismic zones are required to provide proper stiffness and load-bearing capacity to resist frequent earthquakes, and possess proper ductility and energy-dissipating capacity to prevent collapse under rare earthquakes. To meet these requirements, the concept of structural energy-dissipation techniques for the bi-functions of load-bearing and energy dissipating are proposed. A number of structural metal energy-dissipation elements, such as buckling-restrained steel plate shear walls, non-buckling corrugated steel plate shear walls, two-level yielding steel coupling beams and energy-dissipative columns, have been developed. They are designed to provide stiffness/strength to guarantee the operation of buildings under frequent earthquakes, but also dissipate energy to reduce seismic effects to a considerable extent for collapse-prevention of buildings. The experimental and theoretical studies on these structural metal energy-dissipating dampers are presented. The efficiency of these structural dampers for disaster mitigation of buildings against earthquakes are also presented to provide a reference for their practical application.

Photonic crystal fibers (PCFs) made from ZBLAN glass are of great interest for generating broadband supercontinua extending into the ultraviolet and mid-infrared regions. Precise sub-micrometer structuring makes it possible to adjust the modal dispersion over a wide range, making the generation of new frequencies more efficient. Here we report a novel ZBLAN PCF with six cores, each containing a central nanobore of a diameter ∼330  nm. Each nanobore core supports several guided modes, and the presence of the nanobore significantly modifies the dispersion, strongly influencing the dynamics and the extent of supercontinuum generation. Spectral broadening is observed when a single core is pumped in the fundamental and first higher order core modes with 200 fs long pulses at a wavelength of 1042 nm. Frequency-resolved optical gating is used to characterize the output pulses when pumping in the lowest order mode. The results are verified by numerical simulations.

The study is based on the top papers included in the Essential Science Indicators during covering the period of from 2011 to 2015. The results of this study convincingly contribute to a comprehensive comparison of eight world-famous technological institutions; Imperial College, London (ICL), University of California, Berkeley(UCB), California Institute of Technology(CIT), Georgia Institute of Technology(GIT), Tsinghua University(THU), Shanghai Jiao Tong University(SGTU), Xìan Jiao Tong University(SCU), and Harbin Institute of Technology(HIT). Extensive quantitative sources, including total citations of highly cited papers and the average number of references per article are tested by Gini coefficient, mean value and Z score models so as to analyze academic discipline structures and academic influences, as well as put forwards the “Dual Objectives” ( top disciplines and top universities).

We observed narrow-linewidth lasing spikes in a coherent Brillouin random fiber laser, which uses a continuous-wave fiber laser as the pump source. In this random laser configuration, stimulated Brillouin scattering in a section of conventional single-mode fiber acts as the gain mechanism, while Rayleigh scattering in another section of the nonuniform fiber provides randomly distributed feedback. Above the lasing threshold, coherent random lasing spikes, with a linewidth of ∼10  Hz, are measured at the top of the Brillouin gain spectrum. It results from high quality coherent resonance of the Brillouin Stokes light in this open laser cavity.

We report a distributed optical fiber birefringence measurement method based on homodyne Brillouin optical time-domain reflectometry (BOTDR). Unlike conventional BOTDR, which requires scanning of the local oscillator to get the Brillouin spectrum, instead we propose the beat period measurement of fast and slow components of the backscattered Brillouin signal in single-mode fibers using homodyne detection. The beat period is measured by detecting the envelope of the Brillouin beat signal, which gives the beat length at different fiber locations, so that birefringence can be calculated accordingly. The distributed birefringence of a 1.7 km SMF-28 and a 4.3 km large-effective-area fiber were measured with 0.6 m spatial resolution without frequency scanning of the Brillouin spectrum.

We report a single frequency lasing phenomenon with a narrow linewidth of ∼3  kHz in cascaded fiber that is composed of three types of low-loss communication fibers. The Rayleigh scattering of the Brillouin Stokes light created in the middle fiber section along both directions is enhanced by the other two fiber sections. When the Brillouin gain of the middle fiber exceeds the effective loss of the Brillouin stokes light in a roundtrip, a narrow linewidth lasing is observed on the top of the Brillouin spectrum line of the middle fiber. To the best of our knowledge, it is the first report on Rayleigh scattering-assisted Brillouin lasing with single frequency and narrow linewidth in cascaded low-loss communication fibers.