Feature Fusion Technique Research Articles

Cell Nuclear Segmentation of B-ALL Images Based on MSFF-SegNeXt.

The diagnosis and treatment of B-Lineage Acute Lymphoblastic Leukemia (B-ALL) typically rely on cytomorphologic analysis of bone marrow smears. However, traditional morphological analysis methods require manual operation, leading to challenges such as high subjectivity and low efficiency. Accurate segmentation of individual cell nuclei is crucial for obtaining detailed morphological characterization data, thereby improving the objectivity and consistency of diagnoses. To enhance the accuracy of nucleus segmentation of lymphoblastoid cells in B-ALL bone marrow smear images, the Multi-scale Feature Fusion-SegNeXt (MSFF-SegNeXt) model is hereby proposed, building upon the SegNeXt framework. This model introduces a novel multi-scale feature fusion technique that effectively integrates edge feature maps with feature representations across different scales. Integrating the Edge-Guided Attention (EGA) module in the decoder further enhances the segmentation process by focusing on intricate edge details. Additionally, Hamburger structures are strategically incorporated at various stages of the network to enhance feature expression. These combined innovations enable MSFF-SegNeXt to achieve superior segmentation performance on the SN-AM dataset, as evidenced by an accuracy of 0.9659 and a Dice coefficient of 0.9422. The results show that MSFF-SegNeXt outperforms existing models in managing the complexities of cell nucleus segmentation, particularly in capturing detailed edge structures. This advancement offers a robust and reliable solution for subsequent morphological analysis of B-ALL single cells.

Advanced Point Cloud Techniques for Improved 3D Object Detection: A Study on DBSCAN, Attention, and Downsampling

To address the challenges of limited detection precision and insufficient segmentation of small to medium-sized objects in dynamic and complex scenarios, such as the dense intermingling of pedestrians, vehicles, and various obstacles in urban environments, we propose an enhanced methodology. Firstly, we integrated a point cloud processing module utilizing the DBSCAN clustering algorithm to effectively segment and extract critical features from the point cloud data. Secondly, we introduced a fusion attention mechanism that significantly improves the network’s capability to capture both global and local features, thereby enhancing object detection performance in complex environments. Finally, we incorporated a CSPNet downsampling module, which substantially boosts the network’s overall performance and processing speed while reducing computational costs through advanced feature map segmentation and fusion techniques. The proposed method was evaluated using the KITTI dataset. Under moderate difficulty, the BEV mAP for detecting cars, pedestrians, and cyclists achieved 87.74%, 55.07%, and 67.78%, reflecting improvements of 1.64%, 5.84%, and 5.53% over PointPillars. For 3D mAP, the detection accuracy for cars, pedestrians, and cyclists reached 77.90%, 49.22%, and 62.10%, with improvements of 2.91%, 5.69%, and 3.03% compared to PointPillars.

Research on Key Technologies of Image Steganography Based on Simultaneous Deception of Vision and Deep Learning Models

As machine learning continues to evolve, traditional image steganography techniques find themselves increasingly unable to meet the dual challenge of deceiving both the human visual system and machine learning models. In response, this paper introduces the Visually Robust Image Steganography (VRIS) model, specifically tailored for this dual deception task. The VRIS model elevates visual security through the use of specialized feature extraction and processing methodologies. It meticulously conducts feature-level fusion between secret images and cover images, ensuring a high level of visual similarity between them. To effectively mislead machine learning models, the VRIS model incorporates a sophisticated strategy utilizing random noise factors and discriminators. This involves adding controlled amounts of random Gaussian noise to the encrypted image, thereby enhancing the difficulty for machine learning frameworks to recognize it. Furthermore, the discriminator is trained to discern between the noise-infused encrypted image and the original cover image. Through adversarial training, the discriminator and VRIS model refine each other, successfully deceiving the machine learning systems. Additionally, the VRIS model presents an innovative method for extracting and reconstructing secret images. This approach safeguards secret information from unauthorized access while enabling legitimate users to non-destructively extract and reconstruct images by leveraging multi-scale features from the encrypted image, combined with advanced feature fusion and reconstruction techniques. Experimental results validate the effectiveness of the VRIS model, achieving high PSNR and SSIM scores on the LFW dataset and demonstrating significant deception capabilities against the ResNet50 model on the Mini-ImageNet dataset, with an impressive misclassification rate of 99.24%.

RSG-YOLOV8: Detection of rice seed germination rate based on enhanced YOLOv8 and multi-scale attention feature fusion.

The lack of obvious difference between germinated seeds and non-germinated seeds will cause the low accuracy of detecting rice seed germination rate, remains a challenging issue in the field. In view of this, a new model named Rice Seed Germination-YOLOV8 (RSG-YOLOV8) is proposed in this paper. This model initially incorporates CSPDenseNet to streamline computational processes while preserving accuracy. Furthermore, the BRA, a dynamic and sparse attention mechanism is integrated to highlight critical features while minimizing redundancy. The third advancement is the employment of a structured feature fusion network, based on GFPN, aiming to reconfigure the original Neck component of YOLOv8, thus enabling efficient feature fusion across varying levels. An additional detection head is introduced, improving detection performance through the integration of variable anchor box scales and the optimization of regression losses. This paper also explores the influence of various attention mechanisms, feature fusion techniques, and detection head architectures on the precision of rice seed germination rate detection. Experimental results indicate that RSG-YOLOV8 achieves a mAP50 of 0.981, marking a 4% enhancement over the mAP50 of YOLOv8 and setting a new benchmark on the RiceSeedGermination dataset for the detection of rice seed germination rate.

Differential evolution optimization based ensemble framework for accurate cervical cancer diagnosis

Cervical cancer remains a significant global health concern, demanding accurate and efficient diagnostic approaches for early detection and treatment. However, there are situations where a single deep learning model might not be sufficient to adequately capture the pertinent information required for precise disease prediction from intricate datasets. In this paper, we introduce an innovative ensemble learning framework that leverages the Differential Evolution (DE) algorithm to enhance the detection of cervical cancer. Our innovative approach combines the collective strength of a diverse set of base models using feature fusion techniques to achieve precise and robust classification. We have developed three base models, denoted as Fusion 1, Fusion 2, and Fusion 3, where each fusion model incorporates two pre-trained models using a feature fusion technique. To capture sequential dependencies and improve feature attention, we have integrated ConvLSTM layers and Squeeze-and-Excitation (SE) blocks into these base models. Subsequently, we've amalgamated the predictions of each base model using a weighted aggregation scheme, with the weights optimized using the DE algorithm. Our ensemble framework introduces a novel method for dynamic weight adjustment, utilizing advanced fusion techniques to improve cervical cancer detection accuracy. This approach ensures robust diagnostic performance across cancer cells datasets, enhancing patient outcomes and community health. Our evaluation covers two datasets, SIPaKMeD and Mendeley LBC, demonstrating the remarkable performance and robustness of our ensemble model. It consistently outperforms individual base models and even surpasses state-of-the-art methods, achieving an impressive accuracy rate of 98.96 % on the Mendeley LBC dataset and 95.68 % on the SIPaKMeD dataset. The incorporation of ConvLSTM layers, SE blocks, and DE optimization significantly contributes to the model's exceptional accuracy, its ability to generalize effectively, and its adaptability to diverse scenarios. Additionally, we employ Grad-CAM to visualize how the base models focus on crucial regions for accurate predictions, providing insights into the model's decision-making process. Our ensemble model exhibits outstanding Area under the Curve (AUC) values in Receiver Operating Characteristic (ROC) analysis, further confirming its effectiveness. This research represents a significant advancement in cervical cancer detection, underscoring the potential of ensemble learning techniques optimized through DE for enhancing diagnostic accuracy and advancing patient care.

A multi-scale feature extraction and fusion-based model for retinal vessel segmentation in fundus images.

In response to the challenge of low accuracy in retinal vessel segmentation attributed to the minute nature of the vessels, this paper proposes a retinal vessel segmentation model based on an improved U-Net, which combines multi-scale feature extraction and fusion techniques. An improved dilated residual module was first used to replace the original convolutional layer of U-Net, and this module, coupled with a dual attention mechanism and diverse expansion rates, facilitates the extraction of multi-scale vascular features. Moreover, an adaptive feature fusion module was added at the skip connections of the model to improve vessel connectivity. To further optimize network training, a hybrid loss function is employed to mitigate the class imbalance between vessels and the background. Experimental results on the DRIVE dataset and CHASE_DB1 dataset show that the proposed model has an accuracy of 96.27% and 96.96%, sensitivity of 81.32% and 82.59%, and AUC of 98.34% and 98.70%, respectively, demonstrating superior segmentation performance.

RT-Cabi: an Internet of Things based framework for anomaly behavior detection with data correction through edge collaboration and dynamic feature fusion.

The rapid advancement of Internet of Things (IoT) technologies brings forth new security challenges, particularly in anomaly behavior detection in traffic flow. To address these challenges, this study introduces RT-Cabi (Real-Time Cyber-Intelligence Behavioral Anomaly Identifier), an innovative framework for IoT traffic anomaly detection that leverages edge computing to enhance the data processing and analysis capabilities, thereby improving the accuracy and efficiency of anomaly detection. RT-Cabi incorporates an adaptive edge collaboration mechanism, dynamic feature fusion and selection techniques, and optimized lightweight convolutional neural network (CNN) frameworks to address the limitations of traditional models in resource-constrained edge devices. Experiments conducted on two public datasets, Edge-IIoT and UNSW_NB15, demonstrate that RT-Cabi achieves a detection accuracy of 98.45% and 90.94%, respectively, significantly outperforming existing methods. These contributions not only validate the effectiveness of the RT-Cabi model in identifying anomalous behaviors in IoT traffic but also offer new perspectives and technological pathways for future research in IoT security.

A Novel Detection Transformer Framework for Ship Detection in Synthetic Aperture Radar Imagery Using Advanced Feature Fusion and Polarimetric Techniques

Ship detection in synthetic aperture radar (SAR) imagery faces significant challenges due to the limitations of traditional methods, such as convolutional neural network (CNN) and anchor-based matching approaches, which struggle with accurately detecting smaller targets as well as adapting to varying environmental conditions. These methods, relying on either intensity values or single-target characteristics, often fail to enhance the signal-to-clutter ratio (SCR) and are prone to false detections due to environmental factors. To address these issues, a novel framework is introduced that leverages the detection transformer (DETR) model along with advanced feature fusion techniques to enhance ship detection. This feature enhancement DETR (FEDETR) module manages clutter and improves feature extraction through preprocessing techniques such as filtering, denoising, and applying maximum and median pooling with various kernel sizes. Furthermore, it combines metrics like the line spread function (LSF), peak signal-to-noise ratio (PSNR), and F1 score to predict optimal pooling configurations and thus enhance edge sharpness, image fidelity, and detection accuracy. Complementing this, the weighted feature fusion (WFF) module integrates polarimetric SAR (PolSAR) methods such as Pauli decomposition, coherence matrix analysis, and feature volume and helix scattering (Fvh) components decomposition, along with FEDETR attention maps, to provide detailed radar scattering insights that enhance ship response characterization. Finally, by integrating wave polarization properties, the ability to distinguish and characterize targets is augmented, thereby improving SCR and facilitating the detection of weakly scattered targets in SAR imagery. Overall, this new framework significantly boosts DETR’s performance, offering a robust solution for maritime surveillance and security.

TAG-fusion: Two-stage attention guided multi-modal fusion network for semantic segmentation

In the current research, leveraging auxiliary modalities, such as depth information or point cloud information, to improve RGB semantic segmentation has shown significant potential. However, existing methods mainly use convolutional modules for aggregating features from auxiliary modalities, thereby lacking sufficient exploitation of long-range dependencies. Moreover, fusion strategies are typically limited to singular approaches. In this paper, we propose a transformer-based multimodal fusion framework to better utilize auxiliary modalities for enhancing semantic segmentation results. Specifically, we employ a dual-stream architecture for extracting features from RGB and auxiliary modalities, respectively. We incorporate both early fusion and deep feature fusion techniques. At each layer, we introduce mixed attention mechanisms to leverage features from other modalities, guiding and enhancing the current modality's features before propagating them to the subsequent stage of feature extraction. After the extraction of features from different modalities, we employ an enhanced cross-attention mechanism for feature interaction, followed by channel fusion to obtain the final semantic features. Subsequently, we provide separate supervision to the network on the RGB stream, auxiliary stream, and fusion stream to facilitate the learning of representations for different modalities. The experimental results demonstrate that our framework exhibits superior performance across diverse modalities. Specifically, our approach achieves state-of-the-art results on the NYU Depth V2, SUN-RGBD, DELIVER and MFNet datasets.

Indoor occupancy monitoring using environmental feature fusion and semi-supervised machine learning models

Smart buildings optimize energy consumption and occupant comfort through Heating, Ventilation and Air Conditioning, and lighting management. Nevertheless, large venues require data fusion techniques to improve analysis and forecasting. This study aims to evaluate the effectiveness of using different feature fusion techniques, environmental sensors, and semi-supervised learning to estimate indoor occupancy in a 230 m2 office. Using five Internet of Things devices measuring air temperature, relative humidity, and barometric pressure, data was collected for 99 days with 6800 entries (on average) and only 14% labeled. Eight feature selection methods were evaluated along with three supervised and two semi-supervised classification methods. Results indicate that the Chi-squared-based approach for feature fusion outperformed others. Similarly, the semi-supervised Self-Training model achieved better performance than the supervised methods. This research shows that combining semi-supervised learning and data fusion allows for estimating the occupancy level in large indoor spaces with high accuracy and low labeling costs. Highlights This study pioneers in exploring semi-supervised learning and distinct feature fusion methods for estimating indoor occupancy levels in a 230 m 2 open office using only Internet of Things (IoT) environmental sensors (air temperature, relative humidity, and barometric pressure). A comprehensive comparison of statistical methods, feature selection, and dimensionality reduction techniques are conducted to determine their ability to generate robust feature fusion sets. The feature fusion selected through the Chi-squared test stood out with a high accuracy F1-score (average of 0.95) and an average accuracy of 0.99. The Self-Training model reached the best performance from semi-supervised learning, with an average F1-Score of 0.90 and an average accuracy of 0.97, based on a dataset with a large proportion of unlabelled data (16,847 entries) and only 9367 labels. For supervised learning, Random Forest achieved a high accuracy (average of 0.98) and F1-score (average of 0.93) across various feature sets.

An Improved YOLOv8n Algorithm for Steel Surface Defect Detection

Due to the manufacturing process and other aspects, steel surface defects vary in shape and size, which is a difficult point for defect detection. For this reason, this paper proposes a steel surface defect detection algorithm based on YOLOv8n. The algorithm improves the C2f module of Backbone in this network by adding Context Guided Block (CG block), which can effectively extract the local features, surrounding context and global context, and integrate this information to improve the precision and accuracy of detection. Meanwhile, we also change the detection header of YOLOv8 to Detect_ASFF by introducing the Adaptive Spatial Feature Fusion (ASFF) technique. This improvement effectively filters out the conflicting information, which can allow the model to have a better adaptability at different scales. In this paper, Shape-IoU is adopted as the bounding box loss function to make up for the traditional IoU loss that ignores the intrinsic properties of the bounding box in the calculation process. After optimisation, mAP@0.5 are improved to 67.9% and mAP@0.5~0.95 are improved to 35.1%, which is 2.8% and 1.1% respectively. The improved algorithm in this paper has improved in detection accuracy, which proves the practicality and effectiveness of the algorithm in this paper.

An adaptive learning based aberrance repressed multi-feature integrated correlation filter for Visual Object Tracking (VOT)

Target tracking via Correlation Filter (CF) is a hot research area of computer vision domain, and offers various credible benefits. Existing CF algorithms face challenges when there are target appearance variations due to background noise, scale and illumination changes, occlusion, and fast motion, which severely degrades the overall tracker performance. To get maximum benefits, an object tracker should perform well with the less computational burden in the presence of real time challenging situations. To address this issue, a novel visual object trackeris proposed based on multi feature fusion and adaptive learning technique with aberrance suppression. At first, multiple features i.e., Histogram of gradient (HOG), Color Naming (CN), saliency, and gray level intensities are combined using feature fusion technique. Further, based on the evaluation of final fused response map using Peak-to-Sidelobe Ratio (PSR), an adaptive learning strategy is integrated to improve the learning phase of tracker. Tracking results show that the proposed strategy beats the other modern CF trackers with Distance Precision (DP) scores of 88.2%, 85.9%, and 74.1% and 64.7% over OTB2013, OTB2015, and TempleColor128 and UAV123 datasets respectively.

A Review of Research on Handwritten Chinese Character Recognition with Multi-Feature Fusion

This paper analyzes the progress of handwritten Chinese character recognition technology, from two perspectives: traditional recognition methods and deep learning-based recognition methods. Firstly, the complexity of Chinese character recognition is pointed out, including its numerous categories, complex structure, and the problem of similar characters, especially the variability of handwritten Chinese characters. Subsequently, recognition methods based on feature optimization, model optimization, and fusion techniques are highlighted. The fusion studies between feature optimization and model improvement are further explored, and these studies further enhance the recognition effect through complementary advantages. Finally, the article summarizes the current challenges of Chinese character recognition technology, including accuracy improvement, model complexity, and real-time problems, and looks forward to future research directions.

TW-YOLO: An Innovative Blood Cell Detection Model Based on Multi-Scale Feature Fusion.

As deep learning technology has progressed, automated medical image analysis is becoming ever more crucial in clinical diagnosis. However, due to the diversity and complexity of blood cell images, traditional models still exhibit deficiencies in blood cell detection. To address blood cell detection, we developed the TW-YOLO approach, leveraging multi-scale feature fusion techniques. Firstly, traditional CNN (Convolutional Neural Network) convolution has poor recognition capabilities for certain blood cell features, so the RFAConv (Receptive Field Attention Convolution) module was incorporated into the backbone of the model to enhance its capacity to extract geometric characteristics from blood cells. At the same time, utilizing the feature pyramid architecture of YOLO (You Only Look Once), we enhanced the fusion of features at different scales by incorporating the CBAM (Convolutional Block Attention Module) in the detection head and the EMA (Efficient Multi-Scale Attention) module in the neck, thereby improving the recognition ability of blood cells. Additionally, to meet the specific needs of blood cell detection, we designed the PGI-Ghost (Programmable Gradient Information-Ghost) strategy to finely describe the gradient flow throughout the process of extracting features, further improving the model's effectiveness. Experiments on blood cell detection datasets such as BloodCell-Detection-Dataset (BCD) reveal that TW-YOLO outperforms other models by 2%, demonstrating excellent performance in the task of blood cell detection. In addition to advancing blood cell image analysis research, this work offers strong technical support for future automated medical diagnostics.

Efficient crop segmentation net and novel weed detection method

Computer vision-based precision weed control offers a promising avenue for reducing herbicide input and the associated costs of weed management. However, the substantial investments in time and labor required for the collection and annotation of weed image data pose challenges to develop effective deep learning models. The limitation also stems from the challenges in achieving accurate and reliable detection of weeds across varying growth stages, densities, and ecotypes in field scenarios. To address these issues, this research investigated a novel methodology employing a segmentation algorithm to accurately mark the contour information of crops in the image and detect weeds through image processing technology. Furthermore, a novel segmentation network was developed based on the YOLO architecture to address the substantial computing resource demands associated with segmentation algorithms. This was achieved through the design of a new backbone, incorporation of an attention mechanism, and modification of the feature fusion technique. The novel network achieved higher segmentation accuracy with less computational demands. The effectiveness of three different attention modules on segmentation tasks was additionally investigated. Experimental results showed that the insertion of Criss-cross Attention significantly improved the model's performance and was subsequently incorporated into our enhanced methodology. The enhanced model achieved a Mean Intersection over Union (mIoU50) of 90.9 %, with precision increasing by 5.9 % and Giga FLoating-point Operations Per Second (GFLOPs) reduced by 15.56 %, demonstrating enhanced suitability for deployment in resource-constrained computing environments. The findings presented in this study hold substantial theoretical and practical implications for precise weed management.

An ultra lightweight neural network for automatic modulation classification in drone communications

Unmanned aerial vehicle (UAV)-assisted communication based on automatic modulation classification (AMC) technology is considered an effective solution to improve the transmission efficiency of wireless communication systems, as it can adaptively select the most suitable modulation method according to the current communication environment. However, many existing deep learning (DL)-based AMC methods cannot be directly applied to UAV platform with limited computing power and storage space, because of the contradiction between accuracy and efficiency. This paper mainly studies the lightweight of DL-based AMC networks to improve adaptability in resource-constrained scenarios. To address this challenge, we propose an ultra-lightweight neural network (ULNN). This network incorporates a lightweight convolutional structure, attention mechanism, and cross-channel feature fusion technique. Additionally, we introduce data augmentation (DA) based on signal phase offsets during the model training process, aimed at improving the model’s generalization ability and preventing overfitting. Through experimental validation on the public dataset RML2016.10 A, the ULNN we proposed achieves an average precision of 62.83% with only 8815 parameters and reaches a peak classification accuracy of 92.11% at SNR = 10 dB. The experimental results show that ULNN can achieve high recognition accuracy while keeping the model lightweight, and is suitable for UAV platform with limited resources.

Towards accurate diagnosis: exploring knowledge distillation and self-attention in multimodal medical image fusion

ABSTRACT Multimodal medical image fusion aims to aggregate significant information based on the characteristics of medical images from different modalities. Existing research in image fusion faces several major limitations, including a scarcity of paired data, noisy and inconsistent modalities, a lack of contextual relationships, and suboptimal feature extraction and fusion techniques. In response to these challenges, this research proposes a novel adaptive fusion approach. Our knowledge distillation (KD) model extracts informative features from multimodal medical images using various key components. A teacher network is employed to emphasise the suitability and complexity of capturing high-level abstract features. The soft labels are utilised to transfer the knowledge between the teacher network as well as the student network. During student network training, we minimise the divergence between these soft labels. To enhance the adaptive fusion of extracted features from different modalities, we apply a self-attention mechanism. Training this self-attention mechanism minimises the loss function, encouraging attention scores to capture relevant contextual relationships between features. Additionally, a cross-modal consistency module aligns the extracted features to ensure spatial consistency and meaningful fusion. Our adaptive fusion strategy effectively combines features to enhance the diagnostic value and quality of fused images. We employ generator and discriminator architectures for synthesising fused images and distinguishing between real and generated fused images. Comprehensive analysis is conducted on the basis of diverse evaluation measures. Experimental results demonstrate improved fusion outcomes with values of 0.92, 41.58, 7.25, 0.958, 0.759, 0.947, 0.90, 7.05, 0.0726, and 76 s for SSIM, PSNR, FF, VIF, UIQI, FMI, EITF, entropy, RMSE, and execution time, respectively.

Dual vision Transformer-DSUNET with feature fusion for brain tumor segmentation

Brain tumors are one of the leading causes of cancer death; screening early is the best strategy to diagnose and treat brain tumors. Magnetic Resonance Imaging (MRI) is extensively utilized for brain tumor diagnosis; nevertheless, achieving improved accuracy and performance, a critical challenge in most of the previously reported automated medical diagnostics, is a complex problem. The study introduces the Dual Vision Transformer-DSUNET model, which incorporates feature fusion techniques to provide precise and efficient differentiation between brain tumors and other brain regions by leveraging multi-modal MRI data. The impetus for this study arises from the necessity of automating the segmentation process of brain tumors in medical imaging, a critical component in the realms of diagnosis and therapy strategy. The BRATS 2020 dataset is employed to tackle this issue, an extensively utilized dataset for segmenting brain tumors. This dataset encompasses multi-modal MRI images, including T1-weighted, T2-weighted, T1Gd (contrast-enhanced), and FLAIR modalities. The proposed model incorporates the dual vision idea to comprehensively capture the heterogeneous properties of brain tumors across several imaging modalities. Moreover, feature fusion techniques are implemented to augment the amalgamation of data originating from several modalities, enhancing the accuracy and dependability of tumor segmentation. The Dual Vision Transformer-DSUNET model's performance is evaluated using the Dice Coefficient as a prevalent metric for quantifying segmentation accuracy. The results obtained from the experiment exhibit remarkable performance, with Dice Coefficient values of 91.47 % for enhanced tumors, 92.38 % for core tumors, and 90.88 % for edema. The cumulative Dice score for the entirety of the classes is 91.29 %. In addition, the model has a high level of accuracy, roughly 99.93 %, which underscores its durability and efficacy in segmenting brain tumors. Experimental findings demonstrate the integrity of the suggested architecture, which has quickly improved the detection accuracy of many brain diseases.

MFF‐DenseNet: Densely Connected Convolutional Network With Multi‐Scale Feature Fusion for Magnetotelluric Noise Suppression

AbstractMagnetotelluric (MT) is a geophysical technique for detecting subsurface electrical structures. However, MT data collected in areas with frequent human activity often encounter various types of electromagnetic (EM) noise, which can mask or distort the signals we aim to analyze. Over the past decades, data processing methods based on deep learning has become the focus of multiple disciplines. Training neural networks to identify and handle noise has been proven effective in reducing the impact of noise. Therefore, ensuring the neural network accurately learns the noise and signal characteristics during the training is crucial. Against this background, we propose a multi‐scale feature fusion technique based on the densely connected network and apply it to processing MT data. First, we construct a data set resembling the noise in field data and use it to train the network. Leveraging dense connections, we extract feature maps of EM noise from noisy data and utilize Spatial Pyramid Pooling to integrate feature maps of various scales, enabling the network to capture features of the noise precisely. At the same time, we reduce the computation of feature fusion by introducing the Channel‐wise Squeezed Layer to compress the channels of the feature maps. Ultimately, we apply the trained model to the field noisy data. The results of synthetic and field data demonstrate that our method suppresses low‐amplitude and continuous high‐amplitude noise while preserving low‐frequency valuable signal. Apparent resistivity‐phase curves and polarization direction shows a noticeable improvement in the mid and low‐frequency bands with our method.

Therapeutic Exercise Recognition Using a Single UWB Radar with AI-Driven Feature Fusion and ML Techniques in a Real Environment.

Physiotherapy plays a crucial role in the rehabilitation of damaged or defective organs due to injuries or illnesses, often requiring long-term supervision by a physiotherapist in clinical settings or at home. AI-based support systems have been developed to enhance the precision and effectiveness of physiotherapy, particularly during the COVID-19 pandemic. These systems, which include game-based or tele-rehabilitation monitoring using camera-based optical systems like Vicon and Microsoft Kinect, face challenges such as privacy concerns, occlusion, and sensitivity to environmental light. Non-optical sensor alternatives, such as Inertial Movement Units (IMUs), Wi-Fi, ultrasound sensors, and ultrawide band (UWB) radar, have emerged to address these issues. Although IMUs are portable and cost-effective, they suffer from disadvantages like drift over time, limited range, and susceptibility to magnetic interference. In this study, a single UWB radar was utilized to recognize five therapeutic exercises related to the upper limb, performed by 34 male volunteers in a real environment. A novel feature fusion approach was developed to extract distinguishing features for these exercises. Various machine learning methods were applied, with the EnsembleRRGraBoost ensemble method achieving the highest recognition accuracy of 99.45%. The performance of the EnsembleRRGraBoost model was further validated using five-fold cross-validation, maintaining its high accuracy.