Artificial intelligence-assisted security is an important field of research in relation to information security. One of the most important tasks is to distinguish between normal and abnormal network traffic (such as malicious or sudden traffic). Traffic data are usually extremely unbalanced, and this seriously hinders the detection of outliers. Therefore, the identification of outliers in unbalanced datasets has become a key issue. To help solve this challenge, there is increasing interest in focusing on one-class classification methods that train models based on the samples of a single given class. In this paper, long short-term memory (LSTM) is introduced into one-class classification, and one-class LSTM (OC-LSTM) is proposed based on the traditional one-class support vector machine (OC-SVM). In contrast with other hybrid deep learning methods based on auto-encoders, the proposed method is an end-to-end training network that uses a loss function such as the OC-SVM optimization objective for model training. A comprehensive experiment on three large complex network traffic datasets showed that this method is superior to the traditional shallow method and the most advanced deep method. Furthermore, the proposed method can provide an effective reference for anomaly detection research in the field of network security, especially for the application of one-class classification.

We report on a high-efficiency, high-power tandem Ho:YAG single-crystal fiber (SCF) laser in-band pumped by a Tm-doped fiber laser at 1907 nm. In addition to the uniform heat distribution resulting from the large surface-to-volume ratio of this fiber-like thin-crystal rod, the long gain region provided by the tandem layout of two SCFs enables high lasing efficiency and power handling capability. More than 100 W output power is achieved at 2.1 μm, corresponding to a slope efficiency of 70.5% and an optical-to-optical efficiency of 67.6%. To the best of our knowledge, this is the highest output power and efficiency ever reported from SCF lasers in the 2-μm spectral range.

A high-energy and high-average-power pulsed fiber laser has been investigated in a master oscillator power amplifier (MOPA) configuration seeding with a diode laser at a programmed pulse duration of & SIM;250 ns . The fiber amplifier successfully demonstrates the pulse with 21.4 mJ at the repetition rate of 50 kHz and a maximum average output power of 1535 W with a slope efficiency of 81.6% at 250 kHz. To overcome fiber nonlinearities such as stimulated Raman scattering (SRS) and self-phase modulation (SPM), extra-large mode area ytterbium (Yb)doped step-index dual cladding fiber has been utilized as gain fiber in the MOPA laser system. The gain saturation effect in the power amplifier was greatly mitigated by the programmed seed signal. This pulse-shaped MOPA system can provide practical applications in many fields such as laser cleaning, paint stripping, and other applications requiring special pulse shapes.& COPY; 2023 Optica Publishing Group

We report on a 10-watt-level, efficient 2 μm Ho:YAG (0.6 at.% Ho³⁺ doped Y₃Al₅O₁₂) single crystal fiber (SCF) laser in-band pumped by a 1.9 μm laser diode. Benefiting from the high gain of the SCF and the unique propagation features of the pump and laser beams, a maximum output power of 10.77 W is achieved with T _OC = 20%, corresponding to a slope efficiency of 61.8%. The higher efficiency of 64.8% and the output power of 9.8 W are obtained with T _OC = 50%. Compared to the traditional bulk crystal, wave-guiding of the pump light in the SCF can extend the gain region and thus improve the laser efficiency, which provides a simple and effective method to improve the utilization rate of the pump light even for the low-brightness laser diodes.

Medical 3D image reconstruction is an important image processing step in medical image analysis. How to speed up the speed while improving the accuracy in 3D reconstruction is an important issue. To solve this problem, this paper proposes a 3D reconstruction method based on image feature point matching. By improving SIFT, the initial matching of feature points is realized by using the neighborhood voting method, and then the initial matching points are optimized by the improved RANSAC algorithm, and a new SFM reconstruction method is obtained. The experimental results show that the feature matching rate of this algorithm on Fountain data is 95.42% and the matching speed is 4.751 s. It can be seen that this algorithm can shorten the reconstruction time and obtain sparse point clouds with more reasonable distribution and better reconstruction effect.

We propose a B-integral management strategy for manipulating the nonlinear effects by employing a discrete single-crystal fiber (SCF) configuration, enabling direct amplification of 2-μm femtosecond pulses at high repetition rates without additional pulse picking, stretching and compression. The system delivers an average power of more than 56 W at 75.45 MHz with extremely high extraction efficiency (>55%) and near-diffraction-limited beam quality (M² < 1.2). The dynamic evolution of the optical spectra and temporal properties in the power amplifier reveals that detrimental nonlinear effects are largely suppressed due to the low accumulated nonlinear phase shift in the discrete SCF layout. This straightforward, compact and relatively simple approach is expected to open a new route to the amplification of 2-μm ultrashort pulses at MHz and kHz repetition rates to achieve high average/peak powers, thereby offering exciting prospects for applications in modern nonlinear photonics.