A switchable multi-wavelength erbium-doped fiber laser (EDFL) based on parallel dual Lyot filters is experimentally demonstrated in this paper. A new type of parallel-structure fiber filter is designed and its spectrum is simulated. The structure is simple. The spectra of the filter changes when adjusting the polarization states of the polarization controllers (PCs). Due to the characteristics of the filter spectra and wavelength competition, the output wavelengths can be single wavelength, dual wavelengths and three wavelengths. When the output is single wavelength, the wavelength range is tunable with a tuning range of ~13 nm. The wavelength interval of dual wavelengths can be adjusted from 3.03 nm to 17.15 nm, which is a relatively wide range. The interval of three wavelengths also changes to some extent. Moreover, the optical signal-to-noise ratios (OSNRs) of the output wavelengths can up to 47 dB.

A novel individually switchable, widely tunable and interval-adjustable multi-wavelength erbium-doped fiber laser (EDFL) is experimentally demonstrated in this paper. It is based on the cascaded structure of SMF-TMF-SMF (STS) filter and Lyot filter with the assistance of nonlinear polarization rotation (NPR). The homemade STS filter is a comb filter that formed by splicing a section of two-mode fiber (TMF) and two segments of single-mode fiber (SMF) with core-offset structures. Compared with other multimode interferences, the spectrum of STS filter is more uniform. When inserted into the laser cavity, this comb filter functions as the wavelength selector with wavelength interval of ~0.8 nm. The insertion loss of the filter is only ~0.4 dB. By properly adjusting the polarization controllers (PCs) in the laser cavity, individually switchable wavelengths with different flexible lasing states can be obtained, such as tunable adjacent single-, dual-, triple-, quadruple- and quintuple-wavelength outputs. The tuning range can up to 30 nm, from 1530 nm to 1560 nm. When the number of output wavelengths is 1~3, the tuning process can keep the step size of 0.8 nm or 1.6 nm. The non-adjacent dual- and triple-wavelength outputs, as well as adjacent sextuple- and septuple-wavelength outputs can also be achieved. The output characteristics, such as flexible switchability and tunability, are difficult to implement by other methods.

We experimentally demonstrate a multi-wavelength fiber laser based on dual-Sagnac comb filter with LP11 modes output. The filter is used to generate comb spectrum, which is consisted of three optical couplers, two polarization controllers and two segments of polarization maintaining fibers. The proposed fiber laser can stably output up to quintuple-wavelength with the side-mode-suppression ratio higher than 40 dB. Sextuple-wavelength lasing is also obtained, but due to the strong mode competition, the stability should be further improved. A mode selective coupler (MSC) is utilized as mode converter in the cavity. Thus, the multi-wavelength LP11 modes is feasible to generate, owing to the combination of MSC and the filter. The mode purity of the output LP11 modes is higher than 95%. This work has potential applications in optical communication and optical sensing system.

Image-based virtual try-on aims to synthesize the customer image with an in-shop clothes image to acquire seamless and natural try-on results, which have attracted increasing attentions. The main procedures of image-based virtual try-on usually consist of clothes image generation and try-on image synthesis, whereas prior arts cannot guarantee satisfying clothes results when facing large geometric changes and complex clothes patterns, which further deteriorates the afterwards try-on results. To address this issue, we propose a novel virtual try-on network based on landmark-guided shape matching (LM-VTON). Specifically, the clothes image generation progressively learns the warped clothes and refined clothes in an end-to-end manner, where we introduce a landmark-based constraint in Thin-Plate Spline (TPS) warping to inject finer deformation constraints around the clothes. The try-on process synthesizes the warped clothes with personal characteristics via a semantic indicator. Qualitative and quantitative experiments on two public datasets validate the superiority of the proposed method, especially for challenging cases such as large geometric changes and complex clothes patterns. Code will be available at https://github.com/lgqfhwy/LM-VTON.

In this paper, the long short-term memory with dense neural network (LSTM-DNN) is first introduced to calculate marine drifting trajectory. Based on the Internet of Things technology and the LSTM-DNN algorithm, the marine drifting trajectory model is established. In this model, the information such as wind field, temperature field, ocean current motion field, and target attributes are included, and the influences of the above information on the trajectory model are studied in detail. In order to verify the proposed model, the marine experiments are carried out in the end. The results show that the predicted trajectory data matches well with the experimental trajectory data. By introducing DNN into the algorithm, computational accuracy of drifting trajectory can be significantly improved compared with the conventional LSTM-based prediction model. A detailed comparison of the two algorithms has also been given in the paper. The proposed remote sensing of marine drifting trajectory model can provide a high accurate trajectory prediction and will lead an important guidance in the marine search and rescue work.

Millimeter-wave (mmWave) communication is a key technology that enables autonomous vehicles and roadside units to collaborate in environment perception, thereby improving traffic efficiency and safety to a new level. Some recent works have focused on relay-based solutions to overcome inherent defects of mmWave, such as severe path loss and high sensitivity to obstacles. However, they both neglect the selfish behaviors of vehicles. Considering the application-oriented nature of vehicular communication, we propose an information value-based relay strategy in this paper to characterize the selfishness of vehicles, which enables vehicles to make relay decisions through the value assessment of messages. Specifically, an extensible relay probability model is introduced to evaluate the information value based on the required awareness range of vehicles. By using stochastic geometry (SG), we construct an integrated three-dimensional (3D) system model for the vehicular relay-based vehicle-to-infrastructure (V2I) mmWave downlink, and derive the exact expressions of both blockage and outage probabilities. Simulations verify the analytical results and reveal the highest connection reliability of the vehicle exists and relies on both network- and application-related parameters. Moreover, the selfishness of vehicles deteriorates the connection reliability, especially under the conditions of the small information value of forwarded messages. These results laid the foundation for the further expansion of the information value-based relay strategies to a wider range of Internet of Vehicles (IoV) applications.