Search Results (207)

Small object detection in aerial imagery remains challenging due to sparse feature representation, limited spatial resolution, and complex background interference. Current deep learning approaches enhance detection performance through multi-scale feature fusion, leveraging convolutional operations to expand the receptive field or self-attention mechanisms for global context modeling. However, these methods primarily rely on spatial-domain features, while self-attention introduces high computational costs, and conventional fusion strategies (e.g., concatenation or addition) often result in weak feature correlation or boundary misalignment. To address these challenges, we propose a unified spatial-frequency modeling and multi-scale alignment fusion framework, termed USF-DETR, for small object detection. The framework comprises three key modules: the Spatial-Frequency Interaction Backbone (SFIB), the Dual Alignment and Balance Fusion FPN (DABF-FPN), and the Efficient Attention-AIFI (EA-AIFI). The SFIB integrates the Scharr operator for spatial edge and detail extraction and FFT/IFFT for capturing frequency-domain patterns, achieving a balanced fusion of global semantics and local details. The DABF-FPN employs bidirectional geometric alignment and adaptive attention to enhance the significance expression of the target area, suppress background noise, and improve feature asymmetry across scales. The EA-AIFI streamlines the Transformer attention mechanism by removing key-value interactions and encoding query relationships via linear projections, significantly boosting inference speed and contextual modeling. Experiments on the VisDrone and TinyPerson datasets demonstrate the effectiveness of USF-DETR, achieving improvements of 2.3% and 1.4% mAP over baselines, respectively, while balancing accuracy and computational efficiency. The framework outperforms state-of-the-art methods in small object detection. Full article

With the rapid development of electronic business, there has been a surge in incidents of trademark infringement, making it imperative to improve the accuracy of trademark retrieval systems as a key measure to combat such illegal behaviors. Evidently, the textual information encompassed within trademarks substantially influences the precision of search results. Considering the diversity of trademark text and the complexity of its design elements, accurately locating and analyzing this text poses a considerable challenge. Against this background, this research has developed an original self-prompting text removal model, denoted as “Self-prompting Trademark Text Removal Based on Multi-scale Texture Aggregation” (abbreviated as MTF-STTR). This model astutely applies a text detection network to automatically generate the required input cues for the Segment Anything Model (SAM) while incorporating the technological benefits of diffusion models to attain a finer level of trademark text removal. To further elevate the performance of the model, we introduce two innovative architectures to the text detection network: the Integrated Differentiating Feature Pyramid (IDFP) and the Texture Fusion Module (TFM). These mechanisms are capable of efficiently extracting multilevel features and multiscale textual information, which enhances the model’s stability and adaptability in complex scenarios. The experimental validation has demonstrated that the trademark text erasure model designed in this paper achieves a peak signal-to-noise ratio as high as 40.1 dB on the SCUT-Syn dataset, which is an average improvement of 11.3 dB compared with other text erasure models. Furthermore, the text detection network component of the designed model attains an accuracy of up to 89.9% on the CTW1500 dataset, representing an average enhancement of 10 percentage points over other text detection networks. Full article

This paper presents an in-depth study of the automated recognition and geometric information quantification of rebar meshes, proposing a deep learning-based method for rebar mesh detection and segmentation. By constructing a diverse rebar mesh image dataset, an improved Unet-based model was developed, incorporating residual modules to enhance the network’s feature extraction capabilities and training efficiency. The study found that the improved model maintains high segmentation accuracy and robustness even in the presence of complex backgrounds and noise. To achieve the precise measurement of rebar spacing, a rebar intersection detection algorithm based on convolution operations was designed, and the IQR (Interquartile Range) algorithm was applied to remove outliers, ensuring the accuracy and reliability of spacing calculations. The experimental results demonstrate that the proposed model and methods effectively and efficiently accomplish the automated recognition and geometric information extraction of rebar meshes, providing reliable technical support for the automated detection and geometric data analysis of rebar meshes in practical engineering applications. Full article

Seismic coherence attributes are valuable for identifying structural features, but they often face challenges due to significant background noise and non-feature-related stratigraphic discontinuities. To address this, it is necessary to apply attribute conditioning to the coherence to enhance the visibility of these structures. The primary challenge of attribute conditioning lies in finding a concise structural representation that isolates only the true interpretive features while effectively removing noise and stratigraphic interference. In this study, we choose sheet-like structures as this concise structural representation, as faults are typically characterized by their thin and narrow profiles. Inspired by multiscale Hessian-based filtering (MHF) and its application on vascular structure detection, we propose a method called anisotropic multiscale Hessian-based sheet-enhancing filtering (AMHSF). This method is specifically designed to extract and magnify sheet-like structures from noisy coherence images, with a novel enhancement function distinct from those traditionally used in vascular enhancement. The effectiveness of our AMHSF is demonstrated through experiments on both synthetic and real datasets, showcasing its potential to improve the identification of structural features in coherence images. Full article

This study aimed to assess and improve the accuracy of a water leakage detection model proposed in preliminary research. The poor results for water leakage sound (recall) and background noise (specificity) were clarified using countermeasures in accordance with each condition. Additionally, frequency amplification in the range of 500–600 Hz, the attenuation of weak components, and a band-stop filter were used to remove the 50 Hz component and harmonics. Pre-processing was carried out in the form of amplification, with weak noise removed using a band-stop filter. The results showed that the application of the proposed model improved the detection accuracy by 80% at the observation points that initially had poor accuracy. Thus, the proposed method was effective at improving the performance of the Recurrence Plot-Convolutional Neural Network (RP-CNN) model for detecting water leakages. Full article

The inhomogeneity of spectral pixel response is an unavoidable phenomenon in hyperspectral imaging, which is mainly manifested by the existence of inhomogeneity banding noise in the acquired hyperspectral data. It must be carried out to get rid of this type of striped noise since it is frequently uneven and densely distributed, which negatively impacts data processing and application. By analyzing the source of the instrument noise, this work first created a novel non-uniform noise removal method for a spatial dimensional push sweep hyperspectral imaging system. Clean and clear medical hyperspectral brain tumor tissue images were generated by combining scene-based and reference-based non-uniformity correction denoising algorithms, providing a strong basis for further diagnosis and classification. The precise procedure entails gathering the reference dark background image for rectification and the actual medical hyperspectral brain tumor image. The original hyperspectral brain tumor image is then smoothed using a weighted least squares algorithm model embedded with bilateral filtering (BLF-WLS), followed by a calculation and separation of the instrument fixed-mode fringe noise component from the acquired reference dark background image. The purpose of eliminating non-uniform fringe noise is achieved. In comparison to other common image denoising methods, the evaluation is based on the subjective effect and unreferenced image denoising evaluation indices. The approach discussed in this paper, according to the experiments, produces the best results in terms of the subjective effect and unreferenced image denoising evaluation indices (MICV and MNR). The image processed by this method has almost no residual non-uniform noise, the image is clear, and the best visual effect is achieved. It can be concluded that different denoising methods designed for different noises have better denoising effects on hyperspectral images. The non-uniformity denoising method designed in this paper based on a spatial dimension push-sweep hyperspectral imaging system can be widely used. Full article

One of the most important nutrients needed for fruit tree growth is nitrogen. For orchards to get targeted, well-informed nitrogen fertilizer, accurate, large-scale, real-time monitoring, and assessment of nitrogen nutrition is essential. This study examines the Leaf Nitrogen Content (LNC) of the custard apple tree, a noteworthy fruit tree that is extensively grown in China’s Yunnan Province. This study uses an ensemble learning technique based on multiple machine learning algorithms to effectively and precisely monitor the leaf nitrogen content in the tree canopy using multispectral canopy footage of custard apple trees taken via Unmanned Aerial Vehicle (UAV) across different growth phases. First, canopy shadows and background noise from the soil are removed from the UAV imagery by using spectral shadow indices across growth phases. The noise-filtered imagery is then used to extract a number of vegetation indices (VIs) and textural features (TFs). Correlation analysis is then used to determine which features are most pertinent for LNC estimation. A two-layer ensemble model is built to quantitatively estimate leaf nitrogen using the stacking ensemble learning (Stacking) principles. Random Forest (RF), Adaptive Boosting (ADA), Gradient Boosting Decision Trees (GBDT), Linear Regression (LR), and Extremely Randomized Trees (ERT) are among the basis estimators that are integrated in the first layer. By detecting and eliminating redundancy among base estimators, the Least Absolute Shrinkage and Selection Operator regression (Lasso)model used in the second layer improves nitrogen estimation. According to the analysis results, Lasso successfully finds redundant base estimators in the suggested ensemble learning approach, which yields the maximum estimation accuracy for the nitrogen content of custard apple trees’ leaves. With a root mean square error (RMSE) of 0.059 and a mean absolute error (MAE) of 0.193, the coefficient of determination (R²) came to 0. 661. The significant potential of UAV-based ensemble learning techniques for tracking nitrogen nutrition in custard apple leaves is highlighted by this work. Additionally, the approaches investigated might offer insightful information and a point of reference for UAV remote sensing applications in nitrogen nutrition monitoring for other crops. Full article

The targets in the aerial view of UAVs are small, scenes are complex, and background noise is strong. Additionally, the low computational capability of UAVs is challenged when trying to meet the requirements of large neural networks. Therefore, a lightweight object detection algorithm tailored for UAV platforms, called RSG-YOLO, is proposed. The algorithm introduces an attention module constructed with receptive field attention and coordinate attention, which helps reduce background noise interference while improving long-range information dependency. It also introduces and refines a fine-grained downsampling structure to minimize the loss of target information during the downsampling process. A general efficient layer aggregation network enhances the base feature extraction module, improving gradient flow information. Additionally, a detection layer rich in small target information is added, while redundant large object detection layers are removed, achieving a lightweight design while enhancing detection accuracy. Experimental results show that, compared to the baseline algorithm, the improved algorithm increases the P, R, [email protected], and [email protected]:0.95 by 6.9%, 7.2%, 8.4%, 5.8%, respectively, on the VisDrone 2019 dataset, and by 5.7%, 9%, 9.3%, 3.6%, respectively, on the TinyPerson dataset, while reducing the number of parameters by 23.3%. This significantly enhances the model’s detection performance and robustness, making it highly suitable for object detection tasks on low-computing-power UAV platforms. Full article

Ship detection under cloudy and foggy conditions is a significant challenge in remote sensing satellite applications, as cloud cover often reduces contrast between targets and backgrounds. Additionally, ships are small and affected by noise, making them difficult to detect. This paper proposes a Cloud Removal and Target Detection (CRTD) network to detect small ships in images with thin cloud cover. The process begins with a Thin Cloud Removal (TCR) module for image preprocessing. The preprocessed data are then fed into a Small Target Detection (STD) module. To improve target–background contrast, we introduce a Target Enhancement module. The TCR and STD modules are integrated through a dual-stage supervision network, which hierarchically processes the detection task to enhance data quality, minimizing the impact of thin clouds. Experiments on the GaoFen-4 satellite dataset show that the proposed method outperforms existing detectors, achieving an average precision (AP) of 88.9%. Full article

Geiger-mode avalanche photodiode (GM-APD) is a single-photon-detection device characterized by high sensitivity and fast response, which enables it to detect echo signals of distant targets effectively. Given that weak and small targets possess relatively small volumes and occupy only a small number of pixels, relying solely on neighborhood information for target reconstruction proves to be difficult. Furthermore, during long-distance detection, the optical reflection cross-section is small, making signal photons highly susceptible to being submerged by noise. In this paper, a noise fitting and removal algorithm (NFRA) is proposed. This algorithm can detect the position of the echo signal from the photon statistical histogram submerged by noise and facilitate the reconstruction of weak and small targets. To evaluate the NFRA method, this paper establishes an optical detection system for remotely detecting active single-photon weak and small targets based on GM-APD. Taking unmanned aerial vehicles (UAVs) as weak and small targets for detection, this paper compares the target reconstruction effects of the peak-value method and the neighborhood method. It is thereby verified that under the conditions of a 7 km distance and a signal-to-background ratio (SBR) of 0.0044, the NFRA method can effectively detect the weak echo signal of the UAV. Full article

Background: Tuberculosis (TB) is a bacterial disease that mainly affects the lungs, but it can also impact other parts of the body, such as the brain, bones, and kidneys. The disease is caused by a bacterium called Mycobacterium tuberculosis and spreads through the air when an infected person coughs or sneezes. TB can be inactive or active; in its active state, noticeable symptoms appear, and it can be transmitted to others. There are ongoing challenges in fighting TB, including resistance to medications, co-infections, and limited resources in areas heavily affected by the disease. These issues make it challenging to eradicate TB. Objective: Timely and precise diagnosis is essential for effective control, especially since TB often goes undetected and untreated, particularly in remote and under-resourced locations. Chest X-ray (CXR) images are commonly used to diagnose TB. However, difficulties can arise due to unusual findings on X-rays and a shortage of radiologists in high-infection areas. Method: To address these challenges, a computer-aided diagnosis (CAD) system that uses the vision transformer (ViT) technique has been developed to accurately identify TB in CXR images. This innovative hybrid CAD approach combines ViT with Principal Component Analysis (PCA) and machine learning (ML) techniques for TB classification, introducing a new method in this field. In the hybrid CAD system, ViT is used for deep feature extraction as a base model, PCA is used to reduce feature dimensions, and various ML methods are used to classify TB. This system allows for quickly identifying TB, enabling timely medical action and improving patient outcomes. Additionally, it streamlines the diagnostic process, reducing time and costs for patients and lessening the workload on healthcare professionals. The TB chest X-ray dataset was utilized to train and evaluate the proposed CAD system, which underwent pre-processing techniques like resizing, scaling, and noise removal to improve diagnostic accuracy. Results: The performance of our CAD model was assessed against existing models, yielding excellent results. The model achieved remarkable metrics: an average precision of 99.90%, recall of 99.52%, F1-score of 99.71%, accuracy of 99.84%, false negative rate (FNR) of 0.48%, specificity of 99.52%, and negative predictive value (NPV) of 99.90%. Conclusions: This evaluation highlights the superior performance of our model compared to the latest available classifiers. Full article

Background/Objectives: Magnetic resonance spectroscopy (MRS) is a valuable tool for studying metabolic processes in vivo. While numerous quantification methods exist, the advanced method for accurate, robust, and efficient spectral fitting (AMARES) is among the most used. This study introduces pyAMARES, an open-source Python implementation of AMARES, addressing the need for a flexible, user-friendly, and versatile MRS quantification tool within the Python ecosystem. Methods: PyAMARES was developed as a Python library, implementing the AMARES algorithm with additional features such as multiprocessing capabilities and customizable objective functions. The software was validated against established AMARES implementations (OXSA and jMRUI) using both simulated and in vivo MRS data. Monte Carlo simulations were conducted to assess robustness and accuracy across various signal-to-noise ratios and parameter perturbations. Results: PyAMARES utilizes spreadsheet-based prior knowledge and fitting parameter settings, enhancing flexibility and ease of use. It demonstrated comparable performance to existing software in terms of accuracy, precision, and computational efficiency. In addition to conventional AMARES fitting, pyAMARES supports fitting without prior knowledge, frequency-selective AMARES, and metabolite residual removal from mobile macromolecule (MM) spectra. Utilizing multiple CPU cores significantly enhances the performance of pyAMARES. Conclusions: PyAMARES offers a robust, flexible, and user-friendly solution for MRS quantification within the Python ecosystem. Its open-source nature, comprehensive documentation, and integration with popular data science tools enhance reproducibility and collaboration in MRS research. PyAMARES bridges the gap between traditional MRS fitting methods and modern machine learning frameworks, potentially accelerating advancements in metabolic studies and clinical applications. Full article

Maintaining the integrity of urban road pavements is vital for public safety, transportation efficiency, and economic stability. However, aging infrastructure and limited budgets make it challenging to detect subsurface defects that can lead to pavement collapses. Traditional inspection methods are often inadequate for identifying such underground anomalies. Ground Penetrating Radar (GPR), especially dual-polarized array systems, offers a non-destructive, high-resolution solution for subsurface inspection. Despite its potential, effectively detecting and analyzing areas at risk of collapse in urban pavements remains a challenge. This study employed a dual-polarized array GPR system to inspect road pavements in London. The research involved comprehensive field testing, including data acquisition, signal processing, calibration, background noise removal, and 3D migration for enhanced imaging. Additionally, Short-Fourier Transform Spectrum (SFTS) analysis was applied to detect moisture-related anomalies. The results show that dual-polarized GPR systems effectively detect subsurface issues like voids, cracks, and moisture-induced weaknesses. The ability to capture data in multiple polarizations improves resolution and depth, enabling the identification of collapse-prone areas, particularly in regions with moisture infiltration. This study demonstrates the practical value of dual-polarized GPR technology in urban pavement inspection, offering a reliable tool for early detection of subsurface defects and contributing to the longevity and safety of road infrastructure. Full article

Automatic noise-robust speaker identification is essential in various applications, including forensic analysis, e-commerce, smartphones, and security systems. Audio files containing suspect speech often include background noise, as they are typically not recorded in soundproof environments. To this end, we address the challenges of noise robustness and accuracy in speaker identification systems. An ensemble approach is proposed combining two different neural network architectures including an RNN and DNN using softmax. This approach enhances the system’s ability to identify speakers even in noisy environments accurately. Using softmax, we combine voice activity detection (VAD) with a multilayer perceptron (MLP). The VAD component aims to remove noisy frames from the recording. The softmax function addresses these residual traces by assigning a higher probability to the speaker’s voice compared to the noise. We tested our proposed solution on the Kaggle speaker recognition dataset and compared it to two baseline systems. Experimental results show that our approach outperforms the baseline systems, achieving a 3.6% and 5.8% increase in test accuracy. Additionally, we compared the proposed MLP system with Long Short-Term Memory (LSTM) and Bidirectional LSTM (BiLSTM) classifiers. The results demonstrate that the MLP with VAD and softmax outperforms the LSTM by 23.2% and the BiLSTM by 6.6% in test accuracy. Full article

It is difficult to automatically recognize defects using digital image processing methods in X-ray radiographs of lap joints made from plates of unequal thickness. The continuous change in the wall thickness of the lap joint workpiece causes very different gray levels in an X-ray background image. Furthermore, due to the shape and fixturing of the workpiece, the distribution of the weld seam in the radiograph is not vertical which results in an angle between the weld seam and the vertical direction. This makes automatic defect detection and localization difficult. In this paper, a method of X-ray image correction based on invariant moments is presented to solve the problem. In addition, a novel background removal method based on image processing is introduced to reduce the difficulty of defect recognition caused by variations in grayscale. At the same time, an automatic defect detection method combining image noise suppression, image segmentation, and mathematical morphology is adopted. The results show that the proposed method can effectively recognize the gas pores in an automatic welded lap joint of unequal thickness, making it suitable for automatic detection. Full article