To achieve continuous demodulation, high precision, and high resolution in the C-band, this paper designs, simulates and prepares a 30-channel array waveguide grating (AWG) based on a silicon dioxide planar optical circuit for fiber Bragg grating (FBG) interrogation and couples the prepared AWG with a photodetector array using hybrid integration technology. The test results indicate that the AWG has a good transmission spectrum, a 3 dB bandwidth of 2.15 nm, an insertion loss of approximately 3.6–4.2 dB, and crosstalk of less than −30 dB. The FBG interrogation system can achieve continuous demodulation in the dynamic range of 1 521–1 569 nm, and realize continuous demodulation in the C-band with a wavelength resolution of 1 pm and a demodulation accuracy of 5.8 pm. This demodulation method provides an optimization direction for researching FBG interrogation systems based on AWGs.

In this study, we focus on the design and fabrication of a universal one-dimensional photonic crystal (1DPC) frequency division device (FDD), facilitating the separation of optical and radar signals from various sources. Utilizing the forbidden band characteristics of the one-dimensional heterogeneous photonic crystal (HPC), we engineer multi-layer dielectric films composed of ZnS, YbF₃, and Ge. We employ the transfer matrix technique and the frequency domain superposition principle to effectively reflect visible and mid-infrared light, thereby optimizing the radar transmittance performance of convex and concave lenses to facilitate the transmission of specific wavelength bands. The film is prepared using vacuum coating technology on a uniquely shaped quartz substrate. Experiment results reveal that the device exhibits an average reflectance of 0.986 in the visible region and 0.99 in the mid-infrared region, while the average transmittance efficiency in the Ka (32–39 GHz) band exceeds 0.9.

Semiconductor lasers have been widely used, but their most important characteristic, the threshold, has not been sufficiently investigated and remains highly controversial until now. Here, we illustrate this issue using an actual GaAs laser diode (LD). We analyze its electrical behavior, optical power, and electroluminescence (EL) spectrum near the threshold region, then propose that one should distinguish between laser transition points, including population inversion, lasing threshold, and saturation, respectively. Simple methods are provided to characterize the population inversion, the lasing threshold, and the saturation point. Additionally, we discuss the physical significance of these three points.

In this work, a facile hydrothermal method was used to grow ZnO nanoflowers onto an interdigital patterned fluorine-doped tin oxide (FTO) glass substrate, and the crystal structure and morphology of the sample were investigated. The results showed that a great number of regular ZnO nanoflowers grew onto the substrate. Furthermore, the ZnO nanoflowers were used as photosensitive layers, an ultraviolet (UV) photodetector was achieved, and its UV sensing performance was evaluated in detail. The results indicated that this ZnO nanoflowers UV photodetector exhibited good response to 365 nm UV light. Its responsivity and detectivity were up to 135.5 A/W (5 V) and 1.5×10¹³ Jones, respectively, and the response speed was rapid (4.96 s). This work explored a new way for achieving the high-performance nanostructured ZnO UV photodetectors.

We proposed a fast and temperature self-compensation method to detect streptomycin sulfate aqueous solution concentration based on two-dimensional (2D) Nb₂CT_X MXene nanosheets functionalized tilted fiber Bragg grating (N-TFBG). The introduction of 2D Nb₂CT_X MXene nanosheets brought more than 100% increase in sensitivity and exhibited a high sensitivity of 253.8 pm·mg⁻¹·mL⁻¹ and a detection limit of 80 µg/mL in range of 0.2–0.8 mg/mL. Due to high sensitivity and compact structure, the N-TFBG sensor has a faster measure speed than previous method. Additionally, the sensor has simultaneous temperature measurement by detecting the cladding mode and the core mode resonance peak simultaneously. Our research will supply a new research platform for online detection of streptomycin sulfate concentration.

Distributed acoustic sensing (DAS) technology is widely used in seismic monitoring, intrusion detection, and other fields due to its advantages of wide monitoring range and low cost. However, the problems of complex signal processing and large data volume limit its applications. This paper proposes a downsampling method based on short-time Fourier transform, which reduces the length of the time series while retaining high-frequency information. Experiments show that this method improves the efficiency and classification performance of the model, with an F1 value of 0.914 7 on a four-class private dataset and an accuracy of 0.994 4 on a two-class public dataset.

This paper is based on automatically detecting hazardous and non-dangerous objects from terahertz images. First, we trained a neural network to automatically analyze dangerous and non-dangerous items, which can be used for experiment with terahertz images generated by a prototype terahertz video system. The system comprises a terahertz video database of people hiding dangerous and non-dangerous objects under their clothing. Secondly, visual geometry group-19 (VGG-19) is trained on dangerous and non-dangerous objects from our database. After training, the accuracy received was 99.6% for safe items and 85.85% for dangerous items. We tested the network with various categories of objects not included in the training set and found that most were correctly identified as dangerous and nondangerous items. Also, we have identified some of the critical issues that need to be addressed to make this technology more accessible and widely used. Our work can pave the way for future research in this field and help to address the challenges associated with terahertz imaging technologies. The paper describes some preliminary terahertz video surveillance experiments necessary for developing a natural terahertz video surveillance system.

To address challenges in pedestrian detection within dense scenes, including high crowd density, severe occlusion, and overlapping individuals, an improved you only look once (YOLO)-based algorithm is proposed. First, deformable convolutions are employed to replace standard convolutions, enhancing the model’s adaptability to variations in shape and appearance under occlusions. Second, a multi-dimensional attention module is designed to emphasize critical local regions and extract more precise feature information. Lastly, a diagonal difference intersection-over-union (IoU) loss function is introduced, which incorporates a measure of the Euclidean distance difference between the main diagonal points of predicted and ground truth bounding boxes, thereby enhancing detection accuracy and regression performance. Experimental results demonstrate that the enhanced algorithm achieves a mean average precision at IoU=0.5 (mAP50) of 75.1% on the public dense pedestrian dataset WiderPerson, an improvement of 1.8% over the original YOLOv5 model, showcasing superior detection performance.

More accurate segmentation of skin cancers in dermoscopy images is crucial for clinical treatment. However, the prevalence of interfering noise in dermoscopy images poses a challenge to its accurate segmentation. For this reason, this paper proposes an improved GLF-Segformer to improve segmentation. The model adds polarized self-attention (PSA) module and R-convolution and attention fusion module (R-CAFM) to the Segformer’s encoder to enhance the ability to capture local information and facilitate the effective fusion of local and global information. The decoder employs an innovative two-stage hybrid up-sampling to effectively reduce information loss. In addition, a new hybrid loss function is designed to further improve the segmentation accuracy of the model at complex boundaries. The experimental results show that GLF-Segformer achieves 90.73% and 89.85% mean intersection over union (mIoU) on two standard datasets, ISIC2017 and ISIC2018, respectively, and exhibits better segmentation performance compared to other comparison algorithms.

To promote the technology of person re-identification (Re-ID) in intelligent video analysis, a new segmentation method of keypoint-based dynamic region partitioning (KDRP) and an improved adaptive average pooling layer list network (APLNet) are proposed in this work. The KDRP addresses the limitations of traditional stripe segmentation methods avoiding the influence of shooting angles and pedestrian postures. The APLNet integrates the adaptive average pooling layer list (AAPLL) module and the priority circle loss (P-circle loss) to solve the problem of inconsistent size of feature map and promote the model performance respectively. Experimental results on different datasets have validated the effectiveness of the proposed method.

For low-light image enhancement tasks, RAW images surpass RGB images due to their high information content, however, their noise and single-channel nature challenge feature extraction. Existing methods using multi-stage convolutional neural network (CNN) frameworks struggle with global feature extraction, while single-stage CNN-transformer fusions often result in residual noise. To overcome these limitations, this paper introduces a multi-stage RAW image enhancement network combining CNN and transformer. Considering the characteristics inherent to the task, we devised a CNN-based denoising block for the denoising stage and incorporated wavelet information to enhance frequency features. A transformer-based correction block has been designed for the color and white balance recovery stage, with the white balance being adjusted dynamically using a signal-to-noise ratio (SNR) map. With this design, our method outperforms other state-of-the-art models in all metrics on the Sony and Fuji datasets of see-in-the-dark (SID), and achieves optimal structural similarity index measurement (SSIM) on the mono-colored raw (MCR) dataset.