Precision-Driven Semantic Segmentation of Pipe Gallery Diseases Using PipeU-NetX: A Depthwise Separable Convolution Approach

Compared with natural images, remote sensing targets have small and dense target shapes as well as complex target backgrounds. As a result, insufficient detection accuracy and target location cannot be accurately identified. So, this paper proposes the YOLO-extract algorithm based on the YOLOv5 algorithm. Firstly, The YOLO-extract algorithm optimized the model structure of the YOLOv5 algorithm. The YOLO-extract algorithm not only deleted the feature layer and prediction head with poor feature extraction ability but also a new feature extractor with stronger feature extraction ability was integrated into the network. At the same time, YOLO-extract borrowed the idea of residual network to integrate Coordinate Attention into the network. Secondly, the mixed dilated convolution was combined with the redesigned residual structure to enhance the feature and location information extraction ability of the shallow layer of the model and optimize the feature extraction ability of the model for different scale targets. Finally, drawing on the idea of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> -IoU Loss, Focal- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> EIoU Loss was designed to replace CIoU Loss, which makes the model bounding box regression faster and the loss lower. The experimental results on the test data set show that compared with the YOLOv5 algorithm, the YOLO-extract algorithm has a faster convergence speed, reduces the calculation amount by 45.3GFLOPs and the number of parameters by 10.526M, but increases the mAP by 8.1% and the detection speed by 3 times.

In the task of music genre classification, feature extraction and classifier modeling are the two key parts that directly affect the classification accuracy. In the traditional classification method, the feature extraction and classification processes are designed separately. First, the features are extracted manually from the original music signal, and then a reasonable classifier is selected to build a model and classify the extracted features. Although the traditional methods have achieved good results in many classification tasks, the feature extraction process is complex and difficult to achieve, and the features required for different classification tasks need to be specially designed, and the extracted features lack of universality. With the successful application and continuous development of deep learning models in other fields, more and more studies begin to use music spectra as the input of deep learning models to classify music genres. However, so far, the accuracy of existing classification methods based on deep learning is not ideal, so this paper mainly studies a classification method based on deep learning to improve the classification accuracy of music genre classification model. In this paper, a parallel structured deep attention classification model is proposed. By training BRNN, it can automatically learn music features from samples. The linear attention model calculates the attention probability distribution on this feature according to the learned features, and reassigns it to the feature representation. Finally, the classification is realized according to the eigenvectors with different weights. Besides the linear attention model with simple structure, a CNN attention model with stronger learning ability is also designed. In order to verify the feasibility and validity of the model, validation experiments were conducted on two standard datasets, GTZAN and Extended Ballroom. The experimental results show that the classification model based on deep parallel attention mechanism has good classification robustness, and the accuracy of classification based on BRNN and parallel CNN attention model reaches 92.7% on Extended Ballroom data set, which is better than the existing classification method based on deep learning. The validity and feasibility of the classification model are proved.

In this paper, we propose a new efficient superpixel-based feature extraction and fusion method on hyperspectral and LiDAR data. Such important factor that the adjacent pixels belong to the same class with high probability is taken into consideration in our method, which means each superpixel can be regarded as a small region consisting of a number of pixels with similar spectral characteristics. In order to represent each superpixel well, we use our Gabor-wavelet-based feature extraction approach instead of morphological APs. A feature selection and fusion process has also been used to reduce the redundancy among Gabor features and make the fused feature more discriminative. The results on the several real dataset indicate that the proposed method provides state-of-the-art classification results, respectively, even when only few samples, i.e., only three samples per class, are labeled.

Soft robots demonstrate significant advantages in applications within complex environments due to their unique material properties and structural designs. However, they also face challenges in fault diagnosis, such as nonlinearity, time variability, and the difficulty of precise modeling. To address these issues, this paper proposes a fault diagnosis method based on multimodal spatiotemporal features and ensemble learning. First, a sliding-window Kalman filter is utilized to eliminate noise interference from multi-source signals, constructing separate temporal and spatial representation spaces. Subsequently, an adaptive weight strategy for feature fusion is applied to train a heterogeneous decision tree model, followed by a dynamic weighted voting mechanism based on confidence levels to obtain diagnostic results. This method optimizes the feature extraction and fusion process in stages, combined with a dynamic ensemble strategy. Experimental results indicate a significant improvement in diagnostic accuracy and model robustness, achieving precise identification of faults in soft robots.

Accurate image segmentation is crucial in medical imaging for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods often fall short in integrating global and local features in a meaningful way, failing to give sufficient attention to abnormal regions and boundary details in medical images. These limitations hinder the effectiveness of segmentation techniques in clinical settings. To address these issues, we propose a novel deep learning-based approach, MIPC-Net, designed for precise boundary segmentation in medical images. Our approach, inspired by radiologists' working patterns, introduces two distinct modules: 1. Mutual Inclusion of Position and Channel Attention (MIPC) Module: To improve boundary segmentation precision, we present the MIPC module. This module enhances the focus on channel information while extracting position features and vice versa, effectively enhancing the segmentation of boundaries in medical images. 2. Skip-Residue Module: To optimize the restoration of medical images, we introduce Skip-Residue, a global residual connection. This module improves the integration of the encoder and decoder by filtering out irrelevant information and recovering the most crucial information lost during the feature extraction process. We evaluate the performance of MIPC-Net on three publicly accessible datasets: Synapse, ISIC2018-Task, and Segpc. The evaluation uses metrics such as the Dice coefficient (DSC) and Hausdorff Distance (HD). Our ablation study confirms that each module contributes to the overall improvement of segmentation quality. Notably, with the integration of both modules, our model outperforms state-of-the-art methods across all metrics. Specifically, MIPC-Net achieves a 2.23 mm reduction in Hausdorff Distance on the Synapse dataset, highlighting the model's enhanced capability for precise image boundary segmentation. The introduction of the novel MIPC and Skip-Residue modules significantly improves feature extraction accuracy, leading to better boundary recognition in medical image segmentation tasks. Our approach demonstrates substantial improvements over existing methods, as evidenced by the results on benchmark datasets.

Objective. Cardiovascular disease (CVD) is a group of diseases affecting cardiac and blood vessels, and short-axis cardiac magnetic resonance (CMR) images are considered the gold standard for the diagnosis and assessment of CVD. In CMR images, accurate segmentation of cardiac structures (e.g. left ventricle) assists in the parametric quantification of cardiac function. However, the dynamic beating of the heart renders the location of the heart with respect to other tissues difficult to resolve, and the myocardium and its surrounding tissues are similar in grayscale. This makes it challenging to accurately segment the cardiac images. Our goal is to develop a more accurate CMR image segmentation approach. Approach. In this study, we propose a regional perception and multi-scale feature fusion network (RMFNet) for CMR image segmentation. We design two regional perception modules, a window selection transformer (WST) module and a grid extraction transformer (GET) module. The WST module introduces a window selection block to adaptively select the window of interest to perceive information, and a windowed transformer block to enhance global information extraction within each feature window. The WST module enhances the network performance by improving the window of interest. The GET module grids the feature maps to decrease the redundant information in the feature maps and enhances the extraction of latent feature information of the network. The RMFNet further introduces a novel multi-scale feature extraction module to improve the ability to retain detailed information. Main results. The RMFNet is validated with experiments on three cardiac data sets. The results show that the RMFNet outperforms other advanced methods in overall performance. The RMFNet is further validated for generalizability on a multi-organ data set. The results also show that the RMFNet surpasses other comparison methods. Significance. Accurate medical image segmentation can reduce the stress of radiologists and play an important role in image-guided clinical procedures.

Remote sensing image scene classification (RSISC) is an active task in the remote sensing community and has attracted great attention due to its wide applications. Recently, the deep convolutional neural networks (CNNs)-based methods have witnessed a remarkable breakthrough in performance of remote sensing image scene classification. However, the problem that the feature representation is not discriminative enough still exists, which is mainly caused by the characteristic of inter-class similarity and intra-class diversity. In this paper, we propose an efficient end-to-end local-global-fusion feature extraction (LGFFE) network for a more discriminative feature representation. Specifically, global and local features are extracted from channel and spatial dimensions respectively, based on a high-level feature map from deep CNNs. For the local features, a novel recurrent neural network (RNN)-based attention module is first proposed to capture the spatial layout information and context information across different regions. Gated recurrent units (GRUs) is then exploited to generate the important weight of each region by taking a sequence of features from image patches as input. A reweighed regional feature representation can be obtained by focusing on the key region. Then, the final feature representation can be acquired by fusing the local and global features. The whole process of feature extraction and feature fusion can be trained in an end-to-end manner. Finally, extensive experiments have been conducted on four public and widely used datasets and experimental results show that our method LGFFE outperforms baseline methods and achieves state-of-the-art results.

Detecting farmland parcels in high-resolution remote sensing images is challenging in smallholder farming systems in China, characterized by fragmented plots, irregular shapes, and varying scales. To improve detection accuracy in these contexts, this study proposes AAMS-YOLO, a YOLO-based farmland parcel detection model. In the feature extraction stage, the model incorporates an Adaptive Mix Attention (AMA) Block, balancing robust feature extraction with low computational overhead through spatial mixing and Efficient Multi-Scale Attention (EMA). During feature enhancement, to effectively detect targets of different scales, the Attentional Scale Sequence Fusion with P2 network (ASFP2Net) integrates the Triple Feature Encoder (TFE) module and Scale Sequence Feature Fusion (SSFF) module. In the prediction stage, a Multi-Scale Attention Head (MSAHead) enhances adaptability through multi-scale attention mechanisms. Extensive experiments on a custom-built dataset validate AAMS-YOLO's effectiveness, demonstrating notable enhancements over the baseline in mAP0.5 (2.6%) and mAP0.5:0.95 (2.2%) and surpassing other state-of-the-art algorithms. The proposed model excels in detecting small and densely overlapping objects through advanced feature fusion and multi-scale processing strategies.

To solve the insufficient ability of the current Thermal InfraRed (TIR) tracking methods to resist occlusion and interference from similar targets, we propose a TIR tracking method based on efficient global information perception. In order to efficiently obtain the global semantic information of images, we use the Transformer structure for feature extraction and fusion. In the feature extraction process, the Focal Transformer structure is used to improve the efficiency of remote information modeling, which is highly similar to the human attention mechanism. The feature fusion process supplements the relative position encoding to the standard Transformer structure, which allows the model to continuously consider the influence of positional relationships during the learning process. It can also generalize to capture the different positional information for different input sequences. Thus, it makes the Transformer structure model the semantic information contained in images more efficiently. To further improve the tracking accuracy and robustness, the heterogeneous bi-prediction head is utilized in the object prediction process. The fully connected sub-network is responsible for the classification prediction of the foreground or background. The convolutional sub-network is responsible for the regression prediction of the object bounding box. In order to alleviate the contradiction between the vast demand for training data of the Transformer model and the insufficient scale of the TIR tracking dataset, the LaSOT-TIR dataset is generated with the generative adversarial network for network training. Our method achieves the best performance compared with other state-of-the-art trackers on the VOT2015-TIR, VOT2017-TIR, PTB-TIR and LSOTB-TIR datasets, and performs outstandingly especially when dealing with severe occlusion or interference from similar objects.

Rolling bearing fault diagnosis, as a core technical means to ensure the safe operation of rotating machinery, is of vital importance for guaranteeing the continuous and reliable operation of equipment. Addressing the shortcomings of traditional algorithms in local feature extraction, global dependency capturing, and nonlinear feature extraction ability, this paper proposes an intelligent fault diagnosis method based on MVO-VMD and CKCT-Net. The method first optimizes the Variational Mode Decomposition (VMD) using the Multi-Verse Optimizer (MVO) to automatically determine the best K value and avoid relying on manual experience settings. CKCT-Net is designed to integrate complementary mechanisms for capturing finegrained local patterns and modeling long-range dependencies, effectively addressing the difficulty of balancing local details and global context in the feature extraction process. In addition, it enhances nonlinear representation capability to improve adaptability to complex distribution features, and incorporates cross-modal information interaction to enable efficient fusion of heterogeneous features, thereby achieving multilevel feature fusion and information synergy and complementarity among different modalities. Finally, experiments were conducted on public datasets from Case Western Reserve University (CWRU) and Paderborn University (PU). The experimental results show that the method outperforms other baseline methods in fault diagnosis tasks, exhibiting better generalization and robustness.

In an effort to overcome the problem of insufficient network ability of YOLOV8 backbone feature extraction and poor feature fusion ability, a C2f module based on attention mechanism is introduced to carry to a new and higher level the model’s ability to express input features. Aiming at the feature fusion in improved YOLOV8, this paper first modifies the structure of FPN in order to effectively integrate the feature information extracted by backbone, so as to enhance the feature fusion ability of the model. In addition, this paper designs a new feature fusion structure to replace the existing PAN structure, make the feature map information input by FPN more reasonably utilized, so as to improve the overall detection accuracy of the model. The experimental results show that the improved YOLOV8 algorithm performs well on the CUHK dataset [email protected] Improved by 1.7%, and the detection speed remained basically unchanged, with excellent performance on WiderPerson, [email protected] Reaching 91.3%, it can quickly and accurately achieve pedestrian detection. In addition, this paper conducts relevant ablation experiments and comparative experiments to validate the effectiveness of each proposed improvement.

Aiming at the low efficiency of monitoring forest fire by manual inspection and sensor acquisition, and being easily affected by the environment, a forest fire detection algorithm PL-YOLO based on YOLOv5 is proposed. In order to enhance the ability of model feature extraction and effectively fuse feature maps at different scales, this paper proposes two feature extraction and fusion modules. First, we design the Res2-C3 Block to extract more detailed features, reduce the repetitive semantic information in the network and expand the differences between features. Second, the EUASPP module is constructed to achieve effective fusion and stacking of feature maps of different sampling rates at the same level at the cost of adding a small amount of operations. Finally, the loss function is replaced. The PL- YOLO achieved a precision of 89.65% and a mAP of 87.78% on the dataset we construct. The experimental results show that the proposed algorithm achieves a good detection effect on smoke and flame targets in fire scenes, and it is possible to deploy it on a robot platform to realize real-time automatic forest fire monitoring.

This study aims to design an epidemic prevention and control bracelet system that integrates convolutional neural network (CNN). This system can collect and process the user’s physiological index data in real time, especially in the remote physical education scene, and provide learners with immediate physiological index feedback and personalized adaptive training suggestions through accurate human action recognition (HAR) technology. One-dimensional acceleration signal is converted into two-dimensional image, and CNN’s powerful feature extraction and classification ability is used to effectively solve the problem that manual feature extraction is complex and nonlinear features are difficult to capture. By considering the joint action trajectory in the time window, a dynamic Recurrence Plot (RP) is constructed to capture the dynamic changes among joints. To input recursive graph data into CNN, it needs to be converted into image form. In the task of HAR, CNN can automatically learn useful features from images without manually designing features. It can not only effectively extract features from images, but also be directly used in classification tasks. Experimental results show that compared with other algorithms, the proposed RP + CNN model has the best performance in action recognition, with an accuracy of 96.89% and a F1 value of 86.76%. RP captures the dynamic patterns and periodic behaviors in time series by visualizing the repeated appearance of system states over time. The RP + CNN model is used to extract and classify human action features, which significantly improves the accuracy and efficiency of HAR. This innovative method not only simplifies the complex process of traditional manual feature extraction, but also enhances the system’s ability to identify nonlinear and complex action patterns, which provides strong technical support for remote physical education.

We propose a novel machine learning model for classification problems, Deep Twin Support Vector Machine (DTWSVM), which combines TWSVM with deep learning ideas. TWSVM is a successful algorithm for classification problems which seeks two nonparallel hyper planes to make each hyper plane close to one class and far from the other as much as possible. And Deep Learning (DL) models have shown good ability in feature extraction and dimension reduction by constructing multi-layer network. Since the feature extraction in DL can reduce feature dimension while maintain the main information of original inputs, we consider constructing a three layer network which contains input layer, hidden layer and output layer. We use two TWSVMs in the hidden layer to extract features based on the projection principle which is derived from Multi-Layer Perceptron (MLP). The two TWSVMs will get four hyper planes by solving four convex quadratic programs. A new dataset which consists of the extracted features with four feature dimensions is produced from the hidden layer and then we can input it to the main TWSVM of the output layer to make final prediction. Similar as TWSVM, we design linear DTWSVM and nonlinear DTWSVM which have been proved to be very effective in classification problems. In numerical experiments, we have obtained 100% prediction accuracy for several datasets which is state-of-the-art performance absolutely!

To address the challenges in PCB surface defect detection, including high labor costs, low detection efficiency, and insufficient detection accuracy, this study innovatively proposes a PCB surface micro-defect detection network called PCB-YOLOV8X based on the YOLOv8 model. By integrating the C2f-DCNV2 adaptive feature extraction module and the SPPF-LSKA efficient feature enhancement module, the network significantly improves the detection accuracy and efficiency for micro-sized defects. Furthermore, an optimized IWD-CIoU loss function is proposed to more precisely evaluate the matching degree between predicted and ground truth bounding boxes. In comparative experiments with different models, PCB-YOLOV8X achieves optimal performance on key metrics, including mAP@0.5 of 98.44%, Precision of 97.67%, and Recall of 96.34%. Compared with the traditional YOLOv8 model, these represent improvements of 2.76% in mAP@0.5, 2.33% in Precision, and 2.59% in Recall. The proposed PCB-YOLOv8X demonstrates enhanced capability in feature extraction and fusion in PCB surface micro-defect detection tasks, effectively solving issues such as low accuracy, missed detections, and false alarms in existing PCB defect detection models.