Search Results (43)

To address challenges in small object detection within drone aerial imagery—such as sparse feature information, intense background interference, and drastic scale variations—this paper proposes YOLO-SR, a lightweight detection algorithm based on attention enhancement and feature reuse mechanisms. First, we designed the lightweight feature extraction module C2f-SA, which incorporates Shuffle Attention. By integrating channel shuffling and grouped spatial attention mechanisms, this module dynamically enhances edge and texture feature responses for small objects, effectively improving the discriminative power of shallow-level features. Second, the Spatial Pyramid Pooling Attention (SPPC) module captures multi-scale contextual information through spatial pyramid pooling. Combined with dual-path (channel and spatial) attention mechanisms, it optimizes feature representation while significantly suppressing complex background interference. Finally, the detection head employs a decoupled architecture separating classification and regression tasks, supplemented by a dynamic loss weighting strategy to mitigate small object localization inaccuracies. Experimental results on the RGBT-Tiny dataset demonstrate that compared to the baseline model YOLOv5s, our algorithm achieves a 5.3% improvement in precision, a 13.1% increase in recall, and respective gains of 11.5% and 22.3% in mAP_0.5 and mAP_0.75, simultaneously reducing the number of parameters by 42.9% (from 7.0 × 10⁶ to 4.0 × 10⁶) and computational cost by 37.2% (from 60.0 GFLOPs to 37.7 GFLOPs). The comprehensive improvement across multiple metrics validates the superiority of the proposed algorithm in both accuracy and efficiency. Full article

A majority of hyperspectral super-resolution methods aim to enhance the spatial resolution of hyperspectral imaging data (HSI) by integrating high-resolution multispectral imaging data (MSI), leveraging rich spectral information for various geospatial applications. Key challenges include spectral distortions from high-frequency spatial data, high computational complexity, and limited training data, particularly for new-generation sensors with unique noise patterns. In this contribution, we propose a novel parallel patch-wise sparse residual learning (P²SR) algorithm for resolution enhancement based on fusion of HSI and MSI. The proposed method uses multi-decomposition techniques (i.e., Independent component analysis, Non-negative matrix factorization, and 3D wavelet transforms) to extract spatial and spectral features to form a sparse dictionary. The spectral and spatial characteristics of the scene encoded in the dictionary enable reconstruction through a first-order optimization algorithm to ensure an efficient sparse representation. The final spatially enhanced HSI is reconstructed by combining the learned features from low-resolution HSI and applying an MSI-regulated guided filter to enhance spatial fidelity while minimizing artifacts. P²SR is deployable on a high-performance computing (HPC) system with parallel processing, ensuring scalability and computational efficiency for large HSI datasets. Extensive evaluations on three diverse study sites demonstrate that P²SR consistently outperforms traditional and state-of-the-art (SOA) methods in both quantitative metrics and qualitative spatial assessments. Specifically, P²SR achieved the best average PSNR (25.2100) and SAM (12.4542) scores, indicating superior spatio-spectral reconstruction contributing to sharper spatial features, reduced mixed pixels, and enhanced geological features. P²SR also achieved the best average ERGAS (8.9295) and Q2n (0.5156), which suggests better overall fidelity across all bands and perceptual accuracy with the least spectral distortions. Importantly, we show that P²SR preserves critical spectral signatures, such as Fe²⁺ absorption, and improves the detection of fine-scale environmental and geological structures. P²SR’s ability to maintain spectral fidelity while enhancing spatial detail makes it a powerful tool for high-precision remote sensing applications, including mineral mapping, land-use analysis, and environmental monitoring. Full article

An HVDC-GIL system with a conical spacer in a radioactive environment is studied in this work using simulated data on COMSOL^® Multiphysics. Electromagnetic simulations on a 2D model were performed with varying ion-pair generation rates and potential applied to the system. This article explores machine learning methods to derive time to steady state, dark current, gas conductivity, and surface charge density expressions. The focus was on constructing symbolic representations, which could be interpretable and less prone to overfitting, using the symbolic regression (SR) and sparse identification of nonlinear dynamics (SINDy) algorithms. The study successfully derived the intended expressions, demonstrating the power of symbolic regression. Predictions of dark currents in the gas–ground electrode interface reported an absolute error and mean absolute percentage error (MAPE) of 1.04 × ${10}^{-4}$ pA and 0.01%, respectively. The solid–ground electrode interface reported an error of 8.99 × ${10}^{-5}$ pA and MAPE of 0.04%, showing strong agreement with simulation data. Expressions for time to steady state had a test error of approximately 110 h with MAPE of around 3%. Steady-state gas conductivity expression achieved an absolute error of 0.55 log(S/m) and MAPE of 1%. An interpretable equation was created with SINDy to model the time evolution of surface charge density, achieving a root mean squared error of 1.12 nC/m²/s across time-series data. These results demonstrate the capability of SR and SINDy to provide interpretable and computationally efficient alternatives to time-consuming numerical simulations of HVDC systems under radiation conditions. While the model provides useful insights, performance and practical applications of the expressions can improve with more diverse datasets, which might include experimental data in the future. Full article

Transformers have recently achieved significant breakthroughs in various visual tasks. However, these methods often overlook the optimization of interactions between convolution and transformer blocks. Although the basic attention module strengthens the feature selection ability, it is still weak in generating superior quality output. In order to address this challenge, we propose the integration of sub-pixel space and the application of sparse coding theory in the calculation of self-attention. This approach aims to enhance the network’s generation capability, leading to the development of a sparse-activated sub-pixel transformer network (SSTNet). The experimental results show that compared with several state-of-the-art methods, our proposed network can obtain better generation results, improving the sharpness of object edges and the richness of detail texture information in super-resolution generated images. Full article

The extraction of the micro-Doppler (m-D) feature based on time-frequency distribution (TFD) is of great significance for target detection and identification. To improve the feature extraction performance, numerous TFDs have been developed, with the majority falling under Cohen’s class. Nevertheless, these TFDs basically face a trade-off between artifact suppression and energy concentration. The main reason is that each Cohen’s class TFD is constructed by applying the two-dimensional Fourier transform to a kerneled ambiguity function directly, while existing kernels generally attenuate artifacts at the expense of losing valuable information. In this paper, a TFD reconstruction method employing an adaptive short-time kernel (ASTK) is developed in the framework of sparse representation (SR) theory to overcome this trade-off and enhance the m-D feature. Firstly, the task of the optimal kernel is explained from the viewpoint of the instantaneous auto-correlation function (IAF). Secondly, based on the quasi-linear frequency modulation feature of most m-D signals during short-time periods, the distribution rule of the short-time IAF (STIAF) in the ambiguity plane is concluded. Guided by this rule, an ASTK that can effectively remove unwanted artifacts with the least information loss is designed. Finally, an SR-based reconstruction procedure is conducted on the kerneled STIAF to generate an artifact-free TFD with high energy concentration, which can effectively enhance the m-D feature. Experiments using both simulated and real-world m-D signals demonstrate the effectiveness of the proposed method. Full article

Nitrogen content is one of the essential elements in citrus leaves (CL), and many studies have been conducted to determine the nutrient content in CL using hyperspectral technology. To address the key problem that the conventional spectral data-denoising algorithms directly discard high-frequency signals, resulting in missing effective signals, this study proposes a denoising preprocessing algorithm, complete ensemble empirical mode decomposition with adaptive noise joint sparse representation (CEEMDAN–SR), for CL hyperspectral data. For this purpose, 225 sets of fresh CL were collected at the Institute of Fruit Tree Research of the Guangdong Academy of Agricultural Sciences, to measure their elemental nitrogen content and the corresponding hyperspectral data. First, the spectral data were preprocessed using CEEMDAN–SR, Stein’s unbiased risk estimate and the linear expansion of thresholds (SURE–LET), sparse representation (SR), Savitzky–Golay (SG), and the first derivative (FD). Second, feature extraction was carried out using principal component analysis (PCA), uninformative variables elimination (UVE), and the competitive adaptive re-weighted sampling (CARS) algorithm. Finally, partial least squares regression (PLSR), support vector regression (SVR), random forest (RF), and Gaussian process regression (GPR) were used to construct a CL nitrogen prediction model. The results showed that most of the prediction models preprocessed using the CEEMDAN–SR algorithm had better accuracy and robustness. The prediction models based on CEEMDAN–SR preprocessing, PCA feature extraction, and GPR modeling had an $R^2$ of 0.944, $NRMSE$ of 0.057, and $RPD$ of 4.219. The study showed that the CEEMDAN–SR algorithm can be effectively used to denoise CL hyperspectral data and reduce the loss of effective information. The prediction model using the CEEMDAN–SR+PCA+GPR algorithm could accurately obtain the nitrogen content of CL and provide a reference for the accurate fertilization of citrus trees. Full article

Sparsity-based fault diagnosis methods have achieved great success. However, fault classification is still challenging because of neglected potential knowledge. This paper proposes a combined sparse representation deep learning (SR-DEEP) method for rolling bearing fault diagnosis. Firstly, the SR-DEEP method utilizes prior domain knowledge to establish a sparsity-based fault model. Then, based on this model, the corresponding regularization parameter regression networks are trained for different running states, whose core is to explore the latent relationship between the regularization parameters and running states. Subsequently, the performance of the fault classification is improved by embedding the trained regularization parameter regression networks into the sparse representation classification method. This strategy improves the adaptability of the sparse regularization parameter, further improving the performance of the fault classification method. Finally, the applicability of the SR-DEEP method for rolling bearing fault diagnosis is validated with the CWRU platform and QPZZ-II platform, demonstrating that SR-DEEP yields superior accuracies of 100% and 99.20% for diagnosing four and five running states, respectively. Comparative studies show that the SR-DEEP method outperforms four sparse representation methods and seven classical deep learning classification methods in terms of the classification performance. Full article

Multimodal medical image fusion is a fundamental, but challenging, problem in the fields of brain science research and brain disease diagnosis, as it is challenging for sparse representation (SR)-based fusion to characterize activity levels with a single measurement and not lose effective information. In this study, the Kronecker-criterion-based SR framework was applied for medical image fusion with a patch-based activity level, integrating salient features of multiple domains. Inspired by the formation process of vision systems, the spatial saliency was characterized by textural contrast (TC), composed of luminance and orientation contrasts, to promote the participation of more highlighted textural information in the fusion process. As a substitute for the conventional l₁-norm-based sparse saliency, the sum of sparse salient features (SSSF) was used as a metric for promoting the participation of more significant coefficients in the composition of the activity level measurement. The designed activity level measurement was verified to be more conducive to maintaining the integrity and sharpness of detailed information. Various experiments on multiple groups of clinical medical images verified the effectiveness of the proposed fusion method in terms of both visual quality and objective assessment. Furthermore, this study will be helpful for the further detection and segmentation of medical images. Full article

In the principle of lens imaging, when we project a three-dimensional object onto a photosensitive element through a convex lens, the point intersecting the focal plane can show a clear image of the photosensitive element, and the object point far away from the focal plane presents a fuzzy image point. The imaging position is considered to be clear within the limited size of the front and back of the focal plane. Otherwise, the image is considered to be fuzzy. In microscopic scenes, an electron microscope is usually used as the shooting equipment, which can basically eliminate the factors of defocus between the lens and the object. Most of the blur is caused by the shallow depth of field of the microscope, which makes the image defocused. Based on this, this paper analyzes the causes of defocusing in a video microscope and finds out that the shallow depth of field is the main reason, so we choose the corresponding deblurring method: the multi-focus image fusion method. We proposed a new multi-focus image fusion method based on sparse representation (DWT-SR). The operation burden is reduced by decomposing multiple frequency bands, and multi-channel operation is carried out by GPU parallel operation. The running time of the algorithm is further reduced. The results indicate that the DWT-SR algorithm introduced in this paper is higher in contrast and has much more details. It also solves the problem that dictionary training sparse approximation takes a long time. Full article

In high-frequency surface wave radar (HFSWR) systems, clutter is a common phenomenon that causes objects to be submerged. Space-time adaptive processing (STAP), which uses two-dimensional data to increase the degrees of freedom, has recently become a crucial tool for clutter suppression in advanced HFSWR systems. However, in STAP, the pattern is distorted if a clutter component is contained in the main lobe, which leads to errors in estimating the target angle and Doppler frequency. To solve the main-lobe distortion problem, this study developed a clutter-suppression method based on beam reshaping (BR). In this method, clutter components were estimated and maximally suppressed in the side lobe while ensuring that the main lobe remained intact. The results of the proposed algorithm were evaluated by comparison with those of standard STAP and sparse-representation STAP (SR-STAP). Among the tested algorithms, the proposed BR algorithm had the best suppression performance and the most accurate main-lobe peak response, thereby preserving the target angle and Doppler frequency information. The BR algorithm can assist with target detection and tracking despite a background with ionospheric clutter. Full article

The objective of this work is to analyze which features are most important in the recognition of facial expressions. To achieve this, we built a facial expression recognition system that learns from a controlled capture data set. The system uses different representations and combines them from a learned model. We studied the most important features by applying different feature extraction methods for facial expression representation, transforming each obtained representation into a sparse representation (SR) domain, and trained combination models to classify signals, using the extended Cohn–Kanade (CK+), BU-3DFE, and JAFFE data sets for validation. We compared 14 combination methods for 247 possible combinations of eight different feature spaces and obtained the most explanatory features for each facial expression. The results indicate that the LPQ (83%), HOG (82%), and RAW (82%) features are those features most able to improve the classification of expressions and that some features apply specifically to one expression (e.g., RAW for neutral, LPQ for angry and happy, LBP for disgust, and HOG for surprise). Full article

The operation data of a tunnel boring machine (TBM) reflects its geological conditions and working status, which can provide critical references and essential information for TBM designers and operators. However, in practice, operation data may get corrupted due to equipment failures or data management errors. Moreover, the working state of a TBM system usually changes, which results in patterns of operation data that vary comparatively. This paper proposes a denoising approach to process the corrupted data. This approach is combined with low-rank matrix recovery (LRMR) and sparse representation (SR) theory. The classical LRMR model requires that the noise must be sparse, but the sparsity of noise cannot be fully guaranteed. In the proposed model, a weighted nuclear norm is utilized to enhance the sparsity of sparse components, and a constraint of condition number is applied to ensure the stability of the model solution. The approach is coupled with a fuzzy c-means algorithm (FCM) to find the natural partitioning using the TBM operation data as input. The performances of the proposed approach are illustrated through an application to the Shenzhen metro. Experimental results show that the proposed approach performs well in corrupted TBM data denoising. The different excavation status of the TBM recognition accuracy is improved remarkably after denoising. Full article

Space-time adaptive processing (STAP) plays an essential role in clutter suppression and moving target detection in airborne radar systems. The main difficulty is that independent and identically distributed (i.i.d) training samples may not be sufficient to guarantee the performance in the heterogeneous clutter environment. Currently, most sparse recovery/representation (SR) techniques to reduce the requirement of training samples still suffer from high computational complexities. To remedy this problem, a fast group sparse Bayesian learning approach is proposed. Instead of employing all the dictionary atoms, the proposed algorithm identifies the support space of the data and then employs the support space in the sparse Bayesian learning (SBL) algorithm. Moreover, to extend the modified hierarchical model, which can only apply to real-valued signals, the real and imaginary components of the complex-valued signals are treated as two independent real-valued variables. The efficiency of the proposed algorithm is demonstrated both with the simulated and measured data. Full article

Complementary information from multi-sensors can be combined to improve the availability and reliability of stand-alone data. Typically, multispectral (MS) images contain plentiful spectral information of the Earth’s surface that is beneficial for identifying land cover types, while synthetic aperture radar (SAR) images can provide abundant information on the texture and structure of target objects. Therefore, this paper presents a fusion framework to integrate the information from MS and SAR images based on the Laplacian pyramid (LP) and sparse representation (SR) theory. LP is performed to decompose both the multispectral and SAR images into high-frequency components and low-frequency components, so that different processing strategies can be applied to multi-scale information. Low-frequency components are merged based on SR theory, whereas high-frequency components are combined based on a certain activity-level measurement, identifying salient features. Finally, LP reconstruction is performed to obtain the integrated image. We conduct experiments on several datasets to verify the effectiveness of the proposed method. Both visual interpretation and statistical analyses demonstrate that the proposed method strikes a satisfactory balance between spectral information preservation and the enhancement of spatial and textual characteristics. In addition, a further discussion regarding the adjustability property of the proposed method shows its flexibility for further application scenarios. Full article

Acoustic microimaging (AMI), a technology for high-resolution imaging of materials using a scanning acoustic microscope, has been widely used for non-destructive testing and evaluation of electronic packages. Recently, the internal features and defects of electronic packages have reached the resolution limits of conventional time domain or frequency domain AMI methods with the miniaturization of electronic packages. Various time-frequency domain AMI methods have been developed to achieve super-resolution. In this paper, the sparse representation of AMI signals is studied, and a constraint dictionary-based sparse representation (CD-SR) method is proposed. First, the time-frequency parameters of the atom dictionary are constrained according to the AMI signal to constitute a constraint dictionary. Then, the AMI signal is sparsely decomposed using the matching pursuit algorithm, and echoes selection and echoes reconstruction are performed. The performance of CD-SR was quantitatively evaluated by simulated and experimental ultrasonic A-scan signals. The results demonstrated that CD-SR has superior longitudinal resolution and robustness. Full article