Heart rate variability (HRV), as a key indicator for evaluating autonomic nervous system function, has significant value in areas such as cardiovascular disease screening and emotion monitoring. Although traditional contact-based measurement methods offer high precision, they suffer from issues such as poor comfort and low user compliance. This paper proposes a non-contact HRV monitoring method using frequency modulated continuous wave (FMCW) radar, highlighting adaptive cycle segmentation and peak extraction as core innovations. Key advantages of this method include: 1) effective suppression of motion artifacts and respiratory harmonics by leveraging cardiac energy concentration; 2) precise heartbeat cycle identification across physiological states via adaptive segmentation, addressing time-varying differences; 3) adaptive threshold adjustment using discrete energy signals and a support vector machine (SVM) model based on morphological-temporal-spectral characteristics, reducing complexity while maintaining precision. Previous approaches predominantly process radar signals holistically through algorithms to uniformly extract inter-beat intervals (IBIs), which may result in high computational complexity and inadequate dynamic adaptability. In contrast, our method achieved higher precision than conventional holistic processing approaches, while maintaining comparable precision with lower computational complexity than previous optimization algorithms. Experimental results demonstrate that the system achieves an average IBI error of 8.28 ms (RMSE of 15.3 ms), which is reduced by about 66% compared with the traditional holistically peak seeking method. The average errors of SDNN and RMSSD are 2.65 ms and 4.33 ms, respectively. More than 92% of the IBI errors are controlled within 20 ms. The distance adaptability test showed that although the accuracy of long-distance measurement decreased slightly (<6 ms), the overall detection performance remained robust at different distances. This study provided a novel estimation algorithm for non-contact HRV detection, offering new perspectives for future health monitoring.

With the rapid development of flexible electronics, the tactile systems for object recognition are becoming increasingly delicate. This paper presents the design of a tactile glove for object recognition, integrating 243 palm pressure units and 126 finger joint strain units that are implemented by piezoresistive Velostat film. The palm pressure and joint bending strain data from the glove were collected using a two-dimensional resistance array scanning circuit and further converted into tactile images with a resolution of 32×32. To verify the effect of tactile data types on recognition precision, three datasets of tactile images were respectively built by palm pressure data, joint bending strain data, and a tactile data combing of both palm pressure and joint bending strain. An improved residual convolutional neural network (CNN) model, SP-ResNet, was developed by light-weighting ResNet-18 to classify these tactile images. Experimental results show that the data collection method combining palm pressure and joint bending strain demonstrates a 4.33% improvement in recognition precision compared to the best results obtained by using only palm pressure or joint bending strain. The recognition precision of 95.50% for 16 objects can be achieved by the presented tactile glove with SP-ResNet of less computation cost. The presented tactile system can serve as a sensing platform for intelligent prosthetics and robot grippers.

The centroid coordinate serves as a critical control parameter in motion systems, including aircraft, missiles, rockets, and drones, directly influencing their motion dynamics and control performance. Traditional methods for centroid measurement often necessitate custom equipment and specialized positioning devices, leading to high costs and limited accuracy. Here, we present a centroid measurement method that integrates 3D scanning technology, enabling accurate measurement of centroid across various types of objects without the need for specialized positioning fixtures. A theoretical framework for centroid measurement was established, which combined the principle of the multi-point weighing method with 3D scanning technology. The measurement accuracy was evaluated using a designed standard component. Experimental results demonstrate that the discrepancies between the theoretical and the measured centroid of a standard component with various materials and complex shapes in the X, Y, and Z directions are 0.003 mm, 0.009 mm, and 0.105 mm, respectively, yielding a spatial deviation of 0.106 mm. Qualitative verification was conducted through experimental validation of three distinct types. They confirmed the reliability of the proposed method, which allowed for accurate centroid measurements of various products without requiring positioning fixtures. This advancement significantly broadened the applicability and scope of centroid measurement devices, offering new theoretical insights and methodologies for the measurement of complex parts and systems.

For the existing deep learning image restoration methods, the joint guidance of structure and texture information is not considered, which leads to structural disorder and texture blur in the restoration results. A generative adversarial mural inpainting algorithm based on structural and texture hybrid enhancement was proposed. Firstly, the structure guidance branch composed of dynamic convolution cascade was constructed to improve the expression ability of structure features, and the structure information was used to guide the encoder coding to enhance the edge contour information of the coding feature map. Then, the multi-granularity feature extraction module was designed to obtain the texture features of texture guided branches, and the multi-scale texture information was used to guide the decoder to reconstruct and repair, so as to improve the texture consistency of murals. Finally, skip connection was used to promote the feature sharing of structure and texture features, and the spectral-normalized PatchGAN discriminator was used to complete the mural restoration. The digital restoration experiment results of real Dunhuang murals showed that the proposed method was better than the comparison algorithms in both subjective and objective evaluation, and the restoration results were clearer and more natural.

The convolutional neural network (CNN) method based on DeepLabv3+ has some problems in the semantic segmentation task of high-resolution remote sensing images, such as fixed receiving field size of feature extraction, lack of semantic information, high decoder magnification, and insufficient detail retention ability. A hierarchical feature fusion network (HFFNet) was proposed. Firstly, a combination of transformer and CNN architectures was employed for feature extraction from images of varying resolutions. The extracted features were processed independently. Subsequently, the features from the transformer and CNN were fused under the guidance of features from different sources. This fusion process assisted in restoring information more comprehensively during the decoding stage. Furthermore, a spatial channel attention module was designed in the final stage of decoding to refine features and reduce the semantic gap between shallow CNN features and deep decoder features. The experimental results showed that HFFNet had superior performance on UAVid, LoveDA, Potsdam, and Vaihingen datasets, and its cross-linking index was better than DeepLabv3+ and other competing methods, showing strong generalization ability.

The traditional EnFCM (Enhanced fuzzy C-means) algorithm only considers the grey-scale features in image segmentation, resulting in less than satisfactory results when the algorithm is used for remote sensing woodland image segmentation and extraction. An EnFCM remote sensing forest land extraction method based on PCA multi-feature fusion was proposed. Firstly, histogram equalization was applied to improve the image contrast. Secondly, the texture and edge features of the image were extracted, and a multi-feature fused pixel image was generated using the PCA technique. Moreover, the fused feature was used as a feature constraint to measure the difference of pixels instead of a single grey-scale feature. Finally, an improved feature distance metric calculated the similarity between the pixel points and the cluster center to complete the cluster segmentation. The experimental results showed that the error was between 1.5% and 4.0% compared with the forested area counted by experts’ hand-drawing, which could obtain a high accuracy segmentation and extraction result.

In order to suppress the white noise interference in partial discharge (PD) detection and accurately extract the characteristics of local discharge pulse acoustic signal of transformer under strong noise environment, the adaptive separation and denoising of the discharge pulse acoustic signal were analyzed under low signal-to-noise ratio (SNR) environment. Firstly, the optimal decomposition mode number K of the variational mode decomposition (VMD) was determined based on Spearman correlation coefficient, then the reliability of the proposed Spearman-variational mode decomposition (SVMD) method decomposition was verified by simulated signals, and finally the actual discharge pulse acoustic signal was decomposed and denoised based on the Spearman correlation coefficient averaging threshold method to extract the eigenmode function components of the discharge pulse signal. The results showed that SVMD adaptively solved the unknown defects of VMD mode number, and effectively extracted the modal components of complex signals, and successfully realized the denoising of transformer partial discharge acoustic signals. The proposed method effectively removed white noise interference in the partial discharge acoustic signal and obtained a smooth filtered signal. It retained the integrity of the partial discharge signal to the maximum extent and was beneficial to the subsequent research of partial discharge. The improvement of VMD was helpful to promote its wide use in industrial equipment condition inspection.

In order to reduce the error judgment of outliers in vehicle temperature prediction and improve the accuracy of single-station processor prediction data, a Kalman filter multi-information fusion algorithm based on optimized P-Huber weight function was proposed. The algorithm took Kalman filter (KF) as the whole frame, and established the decision threshold based on the confidence level of Chi-square distribution. At the same time, the abnormal error judgment value was constructed by Mahalanobis distance function, and the three segments of Huber weight function were formed. It could improve the accuracy of the interval judgment of outliers, and give a reasonable weight, so as to improve the tracking accuracy of the algorithm. The data values of four important locations in the vehicle obtained after optimized filtering were processed by information fusion. According to theoretical analysis, compared with Kalman filtering algorithm, the proposed algorithm could accurately track the actual temperature in the case of abnormal error, and multi-station data fusion processing could improve the overall fault tolerance of the system. The results showed that the proposed algorithm effectively reduced the interference of abnormal errors on filtering, and the synthetic value of fusion processing was more stable and critical.

To enable optimal navigation for unmanned surface vehicle (USV), we proposed an adaptive hybrid strategy-based sparrow search algorithm (SSA) for efficient and reliable path planning. The proposed method began by enhancing the fitness function to comprehensively account for critical path planning metrics, including path length, turning angle, and navigation safety. To improve search diversity and effectively avoid premature convergence to local optima, chaotic mapping was employed during the population initialization stage, allowing the algorithm to explore a wider solution space from the outset. A reverse inertia weight mechanism was introduced to dynamically balance exploration and exploitation across different iterations. The adaptive adjustment of the inertia weight further improved convergence efficiency and enhanced global optimization performance. In addition, a Cauchy-Gaussian hybrid update strategy was incorporated to inject randomness and variation into the search process, which helped the algorithm escape local minima and maintain a high level of solution diversity. This approach significantly enhanced the robustness and adaptability of the optimization process. Simulation experiments confirmed that the improved SSA consistently outperformed benchmark algorithms such as the original SSA, PSO, and WMR-SSA. Compared with the three algorithms, in the simulated sea area, the path lengths of the proposed algorithm are reduced by 21%, 21%, and 16%, respectively, and under the actual sea simulation conditions, the path lengths are reduced by 13%, 15%, and 11%, respectively. The results highlighted the effectiveness and practicality of the proposed method, providing an effective solution for intelligent and autonomous USV navigation in complex ocean environments.

For short-term PV power prediction, based on interval type-2 Takagi-Sugeno-Kang fuzzy logic systems (IT2 TSK FLS), combined with improved grey wolf optimizer (IGWO) algorithm, an IGWO-IT2 TSK FLS method was proposed. Compared with the type-1 TSK fuzzy logic system method, interval type-2 fuzzy sets could simultaneously model both intra-personal uncertainty and inter-personal uncertainty based on the training of the existing error back propagation (BP) algorithm, and the IGWO algorithm was used for training the model premise and consequent parameters to further improve the predictive performance of the model. By improving the gray wolf optimization algorithm, the early convergence judgment mechanism, nonlinear cosine adjustment strategy, and Levy flight strategy were introduced to improve the convergence speed of the algorithm and avoid the problem of falling into local optimum. The interval type-2 TSK FLS method based on the IGWO algorithm was applied to the real-world photovoltaic power time series forecasting instance. Under the same conditions, it was also compared with different IT2 TSK FLS methods, such as type I TSK FLS method, BP algorithm, genetic algorithm, differential evolution, particle swarm optimization, biogeography optimization, gray wolf optimization, etc. Experimental results showed that the proposed method based on IGWO algorithm outperformed other methods in performance, showing its effectiveness and application potential.

To enhance the quality factor and sensitivity of refractive index sensors, a feedback waveguide slot grating micro-ring resonator was proposed. An air-hole grating structure was introduced based on the slot micro-ring, utilizing the reflection of the grating to achieve the electromagnetic-like induced transparency effect at different wavelengths. The high slope characteristics of the EIT-like effect enabled a higher quality factor and sensitivity. The transmission principle of the structure was analyzed using the transmission matrix method, and the transmission spectrum and mode field distribution were simulated using the finite-difference time-domain (FDTD) method, and the device structure parameters were adjusted for optimization. Simulation results show that the proposed structure achieves an EIT-like effect with a quality factor of 59 267.5. In the analysis of refractive index sensing characteristics, the structure exhibits a sensitivity of 408.57 nm/RIU and a detection limit of 6.23×10^-5 RIU. Therefore, the proposed structure achieved both a high quality factor and refractive index sensitivity, demonstrating excellent sensing performance for applications in environmental monitoring, biomedical fields, and other areas with broad market potential.

For the traditional methods of rotor position estimation for permanent magnet synchronous motor (PMSM), the phase shift caused by the introduction of filter will affect the accuracy of rotor position estimation to some extent. This paper presents an improved rotor position estimation method for high frequency square wave signal injection without filter. Firstly, the traditional method injects high-frequency pulse vibration signals into the estimated shafting, and the proposed method injects high-frequency square wave signals into the estimated shafting to avoid the introduction of filters in the process of extracting rotor position information. Then, the rotor position signal is decoupled in the stationary shafting, and the rotor position error after demodulation is processed by PLL. The system realized the signal processing of rotor position without filter, which improved the convergence speed and estimation precision of rotor position and the dynamic response performance of the system. The simulation results showed that the proposed method had fast convergence speed and small phase delay, and better improved the precision of rotor position detection.

The switch machine is a vital component in the railway system, playing a significant role in ensuring the safe operation of trains. To address the shortcomings of existing fault diagnosis methods for the switch machine and leveraging the strong anti-interference and high sensitivity characteristics of vibration signals, we proposed a VMD-SDP-CNN(Variational mode decomposition-Symmetric dot pattern-Convolutional neural network) fault diagnosis method based on switch machine vibration signals. Firstly, the vibration signal of the switch machine was decomposed by VMD to obtain several intrinsic mode function (IMF) components. Secondly, the SDP method was employed to transform the decomposed IMF components into two-dimensional images, and the issue of one-dimensional signal recognition was transformed into the issue of two-dimensional image recognition. Finally, a CNN was used to realize the fault diagnosis of the switch machine. The experimental results showed that the recognition accuracy of the five actual working conditions of the switch machine using this method was superior to that of typical deep learning and machine learning methods, verifying its practicability and effectiveness.

In order to achieve high precision online prediction of surface roughness during turning process and improve cutting quality, a prediction method of turned surface roughness based on Gramian angular difference field (GADF) of multi-channel signal fusion and multi-scale attention residual network (MA-ResNet) was proposed. Firstly, the multi-channel vibration signals were subdivided into various frequency bands using wavelet packet decomposition, and the sensitive channels were selected for signal fusion by doing correlation analysis between the signals of various frequency bands and the surface roughness. Then the fused signals were converted into pictures using GADF image encoding. Finally, the pictures were inputted into the residual network model combining the parallel dilation convolution and attention module for training and verifying the effectiveness of the model performance. The proposed method has a root mean square error of 0.018 7, a mean absolute error of 0.014 3, and a coefficient of determination of 0.869 4 in predicting the surface roughness, which is close to the actual value. Therefore, the proposed method had good engineering significance for high-precision prediction and was conducive to on-line monitoring of surface quality during workpiece processing.

A fast and accurate homography matrix method for four-wheel positioning detection was presented in the paper. Fewer sensors were required with simpler operation and faster detection. Firstly, eight feature points were extracted by using the target detection algorithm based on the fitting method. Secondly, six feature points were obtained by line fitting-based selection. Thirdly, from the selected six feature points, five points were randomly chosen to minimize the re-projection error. Finally, four points were randomly selected from these five feature points to find the homography matrix, and the other point was back to the homography matrix for verification. The experimental results show that the mean re-projection error is reduced by about 3.41%-4.57% compared with the modified RANSAC (Random sample consensus) algorithm. With the optimized algorithm, the error is reduced by about 12.81%-13.86% compared with the improved RANSAC algorithm. Compared with traditional targets, the average calibration time is reduced by about 26.95%-27.88%. The results indicated that the combination of target algorithm and optimization algorithm could ensure the accuracy and reliability of four-wheel positioning.