Terahertz (THz) waves exhibit distinctive properties, such as high transmittance, pronounced absorption, and minimal photon energy , enabling a wide range of applications in biomedical diagnosis, non-destructive testing, and quality/safety monitoring of food and agricultural products. Consequently, THz-based sensors have garnered increasing attention. However, the design of traditional coupling structures fails to effectively match the high-frequency oscillation of THz waves, resulting in low signal energy transmission efficiency and limiting the performance of THz sensors, while microstructure technology can offer a solution by achieving localized enhancement of the electromagnetic field energy through precise matching of sub-wavelength resonance units with the high-frequency oscillation of THz waves, which significantly improves the sensitivity of THz sensors. This review summarizes the basic principles and research status of various THz sensors based on different microstructures, such as split-ring resonators (SRRs), photonic crystals, waveguide resonators, and surface plasmon resonance. Notably, the rapid development of artificial intelligence, especially deep learning, is increasingly influencing THz sensing technologies with its strengths in signal processing, pattern recognition accuracy, and inverse design. Integrating deep learning with THz sensor design enhances feature extraction from complex signals, improves target identification, and enables intelligent optimization of microstructure parameters for high-performance sensor design and performance prediction. This interdisciplinary approach provides a new pathway to overcome traditional design limitations and advance THz sensor performance.

Photoelectrochemical (PEC) biosensors have drawn growing interest due to their capability to detect biomolecules with the help of generating photocurrent during oxidation reactions, followed by their high sensitivity, minimal background interference, cost-efficiency, and portability. This review provides an extensive summary of the photoactive materials that power PEC biosensor performance. We start by outlining the basic ideas and signal-generating processes of PEC biosensing, highlighting the crucial role of charge-carrier dynamics in photocurrent production. The article's main body thoroughly examines several categories of photoactive materials, such as metal oxides, quantum dots, organic materials, plasmonic nanostructures, and two-dimensional nanomaterials. We go over the special qualities, charge-transfer methods, light-harvesting capacities, and effects on biosensor performance of each material type, all supported by current experimental research. To improve sensitivity and selectivity, we also examine key design techniques, including heterojunction formation, surface functionalization, and hot-electron injection. We also discuss the main issues in PEC biosensors, including interference reduction, biocompatibility, material stability, and reproducibility. Lastly, we discuss future directions, emphasizing new materials, innovative device designs, and potential applications in food safety, environmental monitoring, and point-of-care diagnostics. The goal of this thorough overview is to assist researchers in choosing and creating cutting-edge photoactive materials for high-performance PEC biosensors of the future.

Monocular visual positioning systems are valued for their low cost and straightforward calibration. However, the lack of real-scale information and the complexity of initialization processes limit their application in scenarios requiring accurate absolute positioning. Existing solutions present trade-offs: markerless approaches depend on environmental priors (e.g., fixed camera height constraints) for scale recovery, while marker-based methods typically necessitate that target patterns remain within the camera’s field of view throughout the process. To address these challenges, we propose a 3D target-assisted initialization method that enables scale recovery with just two target images. This modular approach can be seamlessly integrated into monocular simultaneous localization and mapping (SLAM) frameworks. We validated our proposed initialization method through integration with ORB-SLAM3 and semi-direct visual odometry (SVO). Experimental results demonstrated that our method provides real-scale information without compromising real-time performance, making it suitable for applications such as indoor navigation and industrial robot localization, where accurate absolute positioning is essential.

Electrical connectors are core functional components in aerospace electrical systems. Pin retraction may lead to signal transmission interruption and even system failure, directly affecting the reliability of electrical equipment and causing incalculable consequences. We propose a high-precision pin-retraction detection method that integrates binocular stereo vision with a multi-constrained optimization matching algorithm, aiming to achieve universal recognition of pins across different connector models and robust detection of pin retraction in complex scenarios. In this study, the Delaunay triangulation algorithm is employed to eliminate the misidentified pins from the template matching algorithm. Furthermore, the pin recognition rate is enhanced to nearly 99.75%, and the accuracy of pin center positioning is significantly improved by integrating a contour fitting and positioning algorithm for pin points. Subsequently, the binocular matching of pins is achieved by combining probabilistic epipolar constraints with geometric constraints, thereby completing the three-dimensional reconstruction of pin points. The Euclidean distance from the three-dimensional pin points to the reference plane is calculated as the pin retraction amount, enabling the quantitative measurement of pin retraction amount. Through the design of multiple experiments for measuring the pin retraction of different-type electrical connectors and the analysis of the results using the Kullback-Leibler (K-L) divergence, it is demonstrated that the system’s measurement accuracy is superior to 0.05 mm, with an repeatability error of less than 0.035 mm. The effectiveness of the proposed pin-retraction detection method is thus verified, and the detection efficiency over manual operations is greatly enhanced to meet the actual industrial inspection requirements.

The combination of strapdown inertial navigation system (SINS), global navigation satellite system (GNSS), and odometer (ODO) is the most practical and cost-effective way to implement a multi-source fusion automotive navigation system. However, the traditional Kalman filtering (KF) algorithm suffers from the inaccuracy of the system state matrix and the measurement noise covariance matrix during vehicle operation, which leads to a decrease in navigation and positioning accuracy. To solve this problem, a measurement adaptive strong tracking Kalman filter (MA-STKF) algorithm is proposed. The algorithm adopts an asymptotic weighting approach to estimate the measurement covariance array by considering new interest time series being actually filtered, introduces a measurement forgetting factor, perform real-time estimation and correction combines with the decay factor of the strong tracking filter, and takes advantage of the difference between the actual measurement error and the predicted covariance to reset the decay factor, which improves the tracking performance of the algorithm. The proposed algorithm is applied to the SINS/GNSS/ODO integrated navigation system, and simulation and vehicle experiments were conducted, improving the positioning longitude by 52.48% and 30.96%, and the positioning latitude by 63.27% and 37.64%, compared to KF and STKF, respectively.

In oil and gas extraction, ferromagnetic metal casings serve as critical infrastructure to ensure the safety of hydrocarbon transport. However, under high-temperature and high-pressure conditions, casings buried deep underground are prone to deformation, twisting, and even rupture due to erosion and corrosion, potentially leading to significant economic losses and safety hazards. Therefore, regular inspection and maintenance of in-service well casings are essential. Pulsed eddy current testing (PECT) has been widely used for casing defect detection owing to its efficiency, non-contact nature, and rich information content. However, the presence of substantial noise during detection degrades the quality of defect detection images. To address this issue, we investigated image processing techniques for casing defect detection images and proposed an image processing algorithm (BIC) based on bidimensional empirical mode decomposition (BEMD), improved wavelet threshold denoising (IWTD), and contrast limited adaptive histogram equalization (CLAHE). The proposed method first applied BEMD-IWTD for noise suppression in defect detection images, followed by CLAHE for image enhancement. To validate the effectiveness of the method, defect detection experiments were conducted on casings with ring-shaped and local defects, and the acquired images were processed. After being processed with the BIC algorithm, ring-shaped defects of different depths could be effectively distinguished, especially the 1 mm and 2 mm deep defects that were previously affected by noise. In the local defect images, small-sized defects difficult to be identified due to noise interference were successfully recognized, and the defect contrast C_d was significantly improved. The results demonstrate that the proposed BIC algorithm effectively suppresses the noise in defect detection images, enhances the contrast between defects and the background, and improves defect recognition and detection accuracy, providing reliable image processing support for subsequent defect analysis.

Aiming at the problems such as halos, artifacts and incomplete dehazing in hazy image restoring processing, a dehazing algorithm of compensated transmission based on negative haze concentration correction is proposed. First of all, the error mechanism is used to compensate for the transmission of the dark channel prior(DCP), observing the relationships among transmission, depth of field, and haze concentration. A negative haze concentration model is constructed to adaptively correct the transmission of gamma in this study. Finally, the channel difference fusion-based median channel is proposed to correct local atmospheric veil and combined with the atmospheric scattering model to recover haze-free image. The experimental results show that the algorithm solves the problems of halos, artifacts and incomplete dehazing with outstanding details and appropriate brightness.

Due to the absorption and scattering of light in water, underwater images commonly exhibit degradation phenomena such as color cast, low visibility, and blurred details. In response to these issues, we propose a color, detail, contrast, and multi-scale fusion underwater image enhancement algorithm called CDCM. The algorithm first uses a color restoration method based on dark and bright channels to effectively correct color distortion of underwater images and restore their natural color balance. Secondly, utilizing morphological operations to enhance the contour and structural information of objects in the image so as to improve detail representation. In addition, the black eagle optimizer (BEO) is introduced and a new fitness function is designed to adaptively optimize image contrast. In the fusion stage, principal component weights are proposed and combined with other weighting strategies to achieve multi-scale image information fusion, enhancing the contrast while preserving rich textures and details. Experimental results on two real underwater image datasets UIEB and RUIE demonstrate that our method effectively reduces degradation phenomena, with image enhancement by improvements in color fidelity, contrast, and detail clarity compared to the existing methods. In terms of objective indicators, our method is also superior to other relevant methods, such as UCIQE, UIQM, AG, IE, PCQI, etc. Our work contributes to advancing underwater image processing techniques.

This paper presents a resolution reconfigurable two-step successive approximation register analog-to-digital (A/D) converter (ADC) with the pseudo-multiple sampling (PMS) and gain error calibration method for CMOS image sensors. The proposed ADC can be configured with 10-bit, 11-bit and 12-bit by adjusting the number of 10-bit A/D conversions, thereby satisfying various demands in different situations. The PMS method enables the attainment of high-resolution ADC results by summing the conversion outputs of several low-resolution ADCs, thereby reducing the number of unit capacitors and the area of the capacitor array. A compensation technique is proposed to expand the quantization range and improve the effective resolution of the proposed ADC. A calibration method suitable for bottom-plate sampling is proposed, which reduces the gain error between reference voltages. Simulated in a 55 nm process, the proposed ADC in the 12-bit mode achieves a differential nonlinearity of +0.47/-0.50 least significant bit (LSB) and an integral nonlinearity of +0.75/-0.84 LSB at a sampling frequency of 3.497×10⁵ per second with the calibration. The effective number of bits reaches 11.63 bits. The area occupied by a single ADC column is 39.5 µm×119.2 µm and the power consumption is 62.8 µW.

To eliminate the complex interference encountered by pressure sensors in practical applications, we designed and fabricated a piezoresistive pressure sensor featuring wide-temperature-range adaptability to harsh environments and high anti-interference characteristics. A circuit integrating conditioning compensation function with signal conversion function was proposed to compensate and convert pressure signals, and an integrated encapsulated housing was designed and fabricated to connect the pressure sensor chip with the PCB circuit for real-time processing of pressure signals. Its anti-interference performance was primarily reflected in reducing interference to the sensor caused by environmental temperature, voltage noise, and long-distance transmission. The thermal zero drift of the pressure sensor was reduced by 88.95%, and thermal sensitivity drift by 76.17% across the temperature range from -40 ℃ to 105 ℃. When subjected to voltage noise, the signal fluctuation was reduced by 99.7% after circuit processing. When subjected to long-distance transmission, the signal degradation after circuit processing was reduced by 89.9%. The results show that the sensor’s anti-interference performance in complex real-world applications has been enhanced, resulting in more reliable output of the sensor.

The introduction of an aluminum-doped zinc oxide (AZO) buffer layer on a glass substrate has been shown to enhance the performance of Ag/ZnO Schottky photodetectors. To further investigate the correlation between the parameters of AZO buffer layer and ZnO active layer and the performance of Ag/ZnO/AZO/Al photodetectors, we constructed an Ag/ZnO/AZO/Al photodetector based on Silvaco TCAD simulation platform to investigate the effects of AZO layer thickness, ZnO layer thickness, AZO doping concentration, and ZnO doping concentration on the device performance. The simulation results demonstrate that the device achieves better performance when the AZO layer thickness ranges from 0.8 μm to 1.2 μm and the ZnO layer thickness ranges from 0.5 μm to 0.8 μm, with an AZO doping concentration of 1×10¹⁹ cm⁻³ and a ZnO doping concentration of 1×10¹⁶ cm⁻³. Increasing the doping concentration of the AZO buffer layer can enhance the electric field intensity at ZnO-AZO interface, effectively preventing photogenerated holes from approaching ZnO-AZO interface so as to reduce interface recombination; while appropriate ZnO layer thickness and doping concentration can optimize the space charge region width and carrier collection efficiency. Under these optimized conditions, the photodetector reaches its optimum performance with a dark current of 1.1×10⁻⁸ A, a photocurrent of 8.41×10⁻⁶ A, and a responsivity of 0.21 A·W^-1. This research helps reduce the number of experiments and associated costs while preparing low-cost and high-performance ZnO photodetectors.

A comprehensive review of the application status, key technical challenges, and future trends of fiber optic sensing technology applied in space propulsion systems is presented, exploring the feasibility and advantages of replacing traditional electronic sensors with fiber optic sensors in extreme space environments. The fundamental principles of fiber optic sensing technology are analyzed, especially focusing on the mathematical models and operational mechanisms of fiber Bragg grating (FBG) and Fabry-Pérot (F-P) cavity sensors. Furthermore, the latest experimental research and technical solutions are summarized in three typical application scenarios: dynamic strain measurement in cryogenic pipelines, design of intelligent propellant tanks, and temperature distribution monitoring of thermal protection materials in electric propulsion systems. Results demonstrate that packaged FBG sensors can effectively suppress spectral distortion at liquid nitrogen temperatures, enabling accurate strain measurement in small-diameter pipelines; fiber optic sensors embedded in carbon fiber composites can provide real-time structural health and leakage monitoring; and distributed optical frequency domain reflectometry (OFDR) systems can achieve millimeter-level spatial resolution for temperature field monitoring. The discussion identifies remaining technical bottlenecks such as environmental adaptability, packaging techniques, cross-sensitivity, and long-term stability. Future development should focus on integration with smart materials, quantum sensing, on-orbit maintenance, and data-driven decision-making to evolve fiber optic sensing from merely replacing traditional sensors towards enabling intelligent structural systems.

An improved CSYOLOv8 model based on YOLOv8 model is developed specifically for identifying defects in printed circuit board (PCB). Firstly, a composite backbone network is designed to carry out additional feature extraction, which enriches the expression ability of features and enhances the detection accuracy of the model. Secondly, a YOLO-FPN (Feature pyramid network) structure is designed to supplant the original neck network, which enhances the feature fusion ability of the model and improves the detection accuracy of small target objects. Furthermore, to enhance the model’s capability to extract tubular features, dynamic snake convolution is implemented. Finally, MPDIoU loss function is employed to enhance both the convergence rate and the precision of the model. Experiments show that the mAP of the improved model on the PCB defect dataset reaches 96.6%, which is 4.5% higher than that of the YOLOv8 model, and the number of parameters is only 3 256 862, and the average detection speed is 51.8 frames per second, which meets the requirements of detection accuracy and efficiency.

In response to the requirements for high-precision detection and diverse data scenarios in the field of intelligent optical sensing, this research combines whispering gallery mode (WGM) microcavity sensing with machine learning to solve the problems of low spectral information utilization, large random errors, and poor adaptability to data scales in the traditional microcavity sensing. Firstly, the WGM microcavity sensing system is used to collect transmission spectral datasets of different scales. Secondly, a multi-layer perceptron (MLP) deep learning algorithm based on full-spectrum feature mapping is adopted to train and test the datasets through hierarchical feature extraction and nonlinear fitting. The results show that the MLP achieves the test accuracy of 99.95% on large datasets. However, it exhibits poor performance on small datasets. Subsequently, the generalized regression neural network (GRNN) is introduced, leveraging its non-iterative training and strong local feature fitting advantages to optimize the small sample scenarios. The results indicate that the GRNN can achieve a test accuracy of 98.85% on small data sample datasets, improving by 10.29% compared to MLP. Finally, this study quantitatively compares and analyzes the test performance of MLP and GRNN models for five datasets of different scales, clarifying the performance advantages of the two models under different data conditions. This study fully utilizes the characteristics of MLP and GRNN models to achieve high-precision detection under different data scales, providing strong technical support for the application of intelligent optical microcavity sensing technology in various scenarios.

Soft sensor technology has been widely applied in key areas of industrial process monitoring. To address challenges such as strong nonlinearity, complex temporal dependencies, and dynamic system behavior commonly encountered in industrial soft sensor data modeling, we propose a hybrid dynamic modeling method that integrates gated recurrent unit (GRU) with temporal convolutional network-Transformer (TCN-Transformer) architecture. TCN-Transformer module is employed to extract multi-scale temporal patterns and capture long-range dependencies among auxiliary variables, while GRU network processes the historical information of target variables through its gated memory mechanism. The complementary feature representations from both components are summed before being passed into a fully connected layer for prediction. To validate the effectiveness of GRU-TCN-Transformer framework, comprehensive case studies were conducted on two typical industrial processes: the prediction of butane (C₄) concentration in a debutanizer column and the estimation of hydrogen sulfide (H₂S) and sulfur dioxide (SO₂)concentrations in a sulfur recovery unit (SRU). Experimental results demonstrate that the proposed hybrid dynamic modeling method significantly outperforms traditional dynamic modeling methods—convolutional neural network (CNN), long short-term memory (LSTM), and TCN—across multiple evaluation metrics. Specifically, for C₄ concentration estimation, the proposed method reduced root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) by 55.0%, 51.0% and 50.1%, respectively, and improved R² by 2.3% compared to the best-performing TCN-Transformer model. For H₂S estimation, it achieved reductions of 30%, 30.61% and 29.23% in RMSE, MAE, and MAPE, respectively, while increasing R² by 11.09% over the best LSTM-TCN-Transformer model. For SO₂ estimation, the proposed model reduced RMSE, MAE, and MAPE by 7.91%, 9.09% and 9.64%, respectively, with a 0.87% increase in R². These comparative results further confirm the improvements in prediction accuracy, indicating that the proposed model is capable of meeting the stringent requirements of industrial applications.