Ablation Experiments Research Articles

An ultrasound image segmentation method for thyroid nodules based on dual-path attention mechanism-enhanced UNet++

PurposeThis study aims to design an auxiliary segmentation model for thyroid nodules to increase diagnostic accuracy and efficiency, thereby reducing the workload of medical personnel.MethodsThis study proposes a Dual-Path Attention Mechanism (DPAM)-UNet++ model, which can automatically segment thyroid nodules in ultrasound images. Specifically, the model incorporates dual-path attention modules into the skip connections of the UNet++ network to capture global contextual information in feature maps. The model’s performance was evaluated using Intersection over Union (IoU), F1_score, accuracy, etc. Additionally, a new integrated loss function was designed for the DPAM-UNet++ network.ResultsComparative experiments with classical segmentation models revealed that the DPAM-UNet++ model achieved an IoU of 0.7451, an F1_score of 0.8310, an accuracy of 0.9718, a precision of 0.8443, a recall of 0.8702, an Area Under Curve (AUC) of 0.9213, and an HD95 of 35.31. Except for the precision metric, this model outperformed the other models on all the indicators and achieved a segmentation effect that was more similar to that of the ground truth labels. Additionally, ablation experiments verified the effectiveness and necessity of the dual-path attention mechanism and the integrated loss function.ConclusionThe segmentation model proposed in this study can effectively capture global contextual information in ultrasound images and accurately identify the locations of nodule areas. The model yields excellent segmentation results, especially for small and multiple nodules. Additionally, the integrated loss function improves the segmentation of nodule edges, enhancing the model’s accuracy in segmenting edge details.

Vessel Type Recognition Using a Multi-Graph Fusion Method Integrating Vessel Trajectory Sequence and Dependency Relations

In the field of research into vessel type recognition utilizing trajectory data, researchers have primarily concentrated on developing models based on trajectory sequences to extract the relevant information. However, this approach often overlooks the crucial significance of the spatial dependency relationships among trajectory points, posing challenges for comprehensively capturing the intricate features of vessel travel patterns. To address this limitation, our study introduces a novel multi-graph fusion representation method that integrates both trajectory sequences and dependency relationships to optimize the task of vessel type recognition. The proposed method initially extracts the spatiotemporal features and behavioral semantic features from vessel trajectories. By utilizing these behavioral semantic features, the key nodes within the trajectory that exhibit dependencies are identified. Subsequently, graph structures are constructed to represent the intricate dependencies between these nodes and the sequences of trajectory points. These graph structures are then processed through graph convolutional networks (GCNs), which integrate various sources of information within the graphs to obtain behavioral representations of vessel trajectories. Finally, these representations are applied to the task of vessel type recognition for experimental validation. The experimental results indicate that this method significantly enhances vessel type recognition performance when compared to other baseline methods. Additionally, ablation experiments have been conducted to validate the effectiveness of each component of the method. This innovative approach not only delves deeply into the behavioral representations of vessel trajectories but also contributes to advancements in intelligent water traffic control.

UAV detection in complex background with multi-scale feature fusion enhancement and channel-weight matching up-sampling

Abstract The reckless flight of unmanned aerial vehicle (UAV) seriously threatens the public and aviation safety. Due to their small size and unobvious features, it remains a great challenge for the current detection algorithms to detect UAV, especially in complex backgrounds with backlighting. To address these issues, the multiscale feature fusion enhancement strategy and channel-weight matching (CWM) rule are proposed in this paper. A multiscale feature fusion enhancement strategy is presented to capture the multi-scale contextual information, which not only suppresses information conflicts but also enhances feature extraction capabilities. Then, an up-sampling method based on CWM is designed to enhance the sensitivity of small object, which uses different up-sampling techniques based on the importance level of each feature channel. Finally, a feature refinement module for small object is designed to further enhance the characterization of their features. The ablation and comparative experiments are carried out on the self-made UAV dataset. Compared to the original YOLOv5 algorithm, the proposed method shows an increase of 3.6% in mAP0.5 and 2.8% in mAP0.5:0.95, respectively. Moreover, the comparative experiments are implemented on the VisDrone2019 dataset, and the results indicate that the mAP0.5 and mAP0.5:0.95 of the proposed method also increase by 4.2% and 1.6%, respectively.

RADD-CycleGAN: unsupervised reconstruction of high-quality ultrasound image based on CycleGAN with residual attention and dual-domain discrimination

Plane wave (PW) imaging is fast, but limited by poor imaging quality. Coherent PW compounding (CPWC) improves image quality but decrease frame rate. In this study, we propose a modified CycleGAN model that combines a residual attention module with a space-frequency dual-domain discriminator, termed RADD-CycleGAN, to rapidly reconstruct high-quality ultrasound images. To enhance the ability to reconstruct image details, we specially design a process of hybrid dynamic and static channel selection followed by the frequency domain discriminator. The low-quality images are generated by the 3-angle CPWC, while the high-quality images are generated as real images (ground truth) by the 75-angle CPWC. The training set includes unpaired images, whereas the images in the test set are paired to verify the validity and superiority of the proposed model. Finally, we respectively design ablation and comparison experiments to evaluate the model performance. Compared with the basic CycleGAN, our proposed method reaches a better performance, with a 7.8% increase in the peak signal-to-noise ratio and a 22.2% increase in the structural similarity index measure. The experimental results show that our method achieves the best unsupervised reconstruction from low quality images in comparison with several state-of-the-art methods.

Pipeline and Rotating Pump Condition Monitoring Based on Sound Vibration Feature-Level Fusion

The rotating pump of pipelines are susceptible to damage based on extended operations in a complex environment of high temperature and high pressure, which leads to abnormal vibrations and noises. Currently, the method for detecting the conditions of pipelines and rotating pumps primarily involves identifying their abnormal sounds and vibrations. Due to complex background noise, the performance of condition monitoring is unsatisfactory. To overcome this issue, a pipeline and rotating pump condition monitoring method is proposed by extracting and fusing sound and vibration features in different ways. Firstly, a hand-crafted feature set is established from two aspects of sound and vibration. Moreover, a convolutional neural network (CNN)-derived feature set is established based on a one-dimensional CNN (1D CNN). For the hand-crafted and CNN-derived feature sets, a feature selection method is presented for significant features by ranking features according to their importance, which is calculated by ReliefF and the random forest score. Finally, pipeline and rotating pump condition monitoring is applied by fusing the significant sound and vibration features at the feature level. According to the sound and vibration signals obtained from the experimental platform, the proposed method was evaluated, showing an average accuracy of 93.27% for different conditions. The effectiveness and superiority of the proposed method are manifested through comparison and ablation experiments.

A text classification method combining in-domain pre-training and prompt learning for the steel e-commerce industry

Purpose With the development of Web information systems, steel e-commerce platforms have accumulated a large number of quality objection texts. These texts reflect consumer dissatisfaction with the dimensions, appearance and performance of steel products, providing valuable insights for product improvement and consumer decision-making. Currently, mainstream solutions rely on pre-trained models, but their performance on domain-specific data sets and few-shot data sets is not satisfactory. This paper aims to address these challenges by proposing more effective methods for improving model performance on these specialized data sets. Design/methodology/approach This paper presents a method on the basis of in-domain pre-training, bidirectional encoder representation from Transformers (BERT) and prompt learning. Specifically, a domain-specific unsupervised data set is introduced into the BERT model for in-domain pre-training, enabling the model to better understand specific language patterns in the steel e-commerce industry, enhancing the model’s generalization capability; the incorporation of prompt learning into the BERT model enhances attention to sentence context, improving classification performance on few-shot data sets. Findings Through experimental evaluation, this method demonstrates superior performance on the quality objection data set, achieving a Macro-F1 score of 93.32%. Additionally, ablation experiments further validate the significant advantages of in-domain pre-training and prompt learning in enhancing model performance. Originality/value This study clearly demonstrates the value of the new method in improving the classification of quality objection texts for steel products. The findings of this study offer practical insights for product improvement in the steel industry and provide new directions for future research on few-shot learning and domain-specific models, with potential applications in other fields.

Deformable multi-level feature network applied to nucleus segmentation

IntroductionThe nucleus plays a crucial role in medical diagnosis, and accurate nucleus segmentation is essential for disease assessment. However, existing methods have limitations in handling the diversity of nuclei and differences in staining conditions, restricting their practical application.MethodsA novel deformable multi-level feature network (DMFNet) is proposed for nucleus segmentation. This network is based on convolutional neural network and divides feature processing and mask generation into two levels. At the feature level, deformable convolution is used to enhance feature extraction ability, and multi-scale features are integrated through a balanced feature pyramid. At the mask level, a one-stage framework is adopted to directly perform instance segmentation based on location.ResultsExperimental results on the MoNuSeg 2018 dataset show that the mean average precision (mAP) and mean average recall (mAR) of DMFNet reach 37.8% and 47.4% respectively, outperforming many current advanced methods. Ablation experiments verified the effectiveness of each module of the network.DiscussionDMFNet provides an effective solution for nucleus segmentation and has important application value in medical image analysis.

HybridGNN: A Self-Supervised Graph Neural Network for Efficient Maximum Matching in Bipartite Graphs

Solving maximum matching problems in bipartite graphs is critical in fields such as computational biology and social network analysis. This study introduces HybridGNN, a novel Graph Neural Network model designed to efficiently address complex matching problems at scale. HybridGNN leverages a combination of Graph Attention Networks (GATv2), Graph SAGE (SAGEConv), and Graph Isomorphism Networks (GIN) layers to enhance computational efficiency and model performance. Through extensive ablation experiments, we identify that while the SAGEConv layer demonstrates suboptimal performance in terms of accuracy and F1-score, configurations incorporating GATv2 and GIN layers show significant improvements. Specifically, in six-layer GNN architectures, the combinations of GATv2 and GIN layers with ratios of 4:2 and 5:1 yield superior accuracy and F1-score. Therefore, we name these GNN configurations HybridGNN1 and HybridGNN2. Additionally, techniques such as mixed precision training, gradient accumulation, and Jumping Knowledge networks are integrated to further optimize performance. Evaluations on an email communication dataset reveal that HybridGNNs outperform traditional algorithms such as the Hopcroft–Karp algorithm, the Hungarian algorithm, and the Blossom/Edmonds’ algorithm, particularly for large and complex graphs. These findings highlight HybridGNN’s robust capability to solve maximum matching problems in bipartite graphs, making it a powerful tool for analyzing large-scale and intricate graph data. Furthermore, our study aligns with the goals of the Symmetry and Asymmetry Study in Graph Theory special issue by exploring the role of symmetry in bipartite graph structures. By leveraging GNNs, we address the challenges related to symmetry and asymmetry in graph properties, thereby improving the reliability and fault tolerance of complex networks.

Data Decomposition Modeling Based on Improved Dung Beetle Optimization Algorithm for Wind Power Prediction

Accurate wind power forecasting is essential for maintaining the stability of a power system and enhancing scheduling efficiency in the power sector. To enhance prediction accuracy, this paper presents a hybrid wind power prediction model that integrates the improved complementary ensemble empirical mode decomposition (ICEEMDAN), the RIME optimization algorithm (RIME), sample entropy (SE), the improved dung beetle optimization (IDBO) algorithm, the bidirectional long short-term memory (BiLSTM) network, and multi-head attention (MHA). In this model, RIME is utilized to improve the parameters of ICEEMDAN, reducing data decomposition complexity and effectively capturing the original data information. The IDBO algorithm is then utilized to improve the hyperparameters of the MHA-BiLSTM model. The proposed RIME-ICEEMDAN-IDBO-MHA-BiLSTM model is contrasted with ten others in ablation experiments to validate its performance. The experimental findings prove that the proposed model achieves MAPE values of 5.2%, 6.3%, 8.3%, and 5.8% across four datasets, confirming its superior predictive performance and higher accuracy.

Prediction of PM2.5 Concentration by Transformer Model Based on Attention Mechanism

An improved transformer model based on a multi-head attention mechanism was constructed for long-term prediction of PM2.5 concentration. The monitoring data and meteorological data from 12 air monitoring stations in Beijing from March 2013 to December 2016 were collected and used in the transformer model. The Pearson coefficient was used to explore the key factors affecting PM2.5 concentration. A convolutional neural network model （ResNet50） and long short-term memory network model （LSTM） were introduced for comparison and explanatory variance （EVS）, coefficient of determination （R2）, mean square error （MSE）, and mean absolute error （MAE） were selected to evaluate the performance of the model. The Pearson coefficient results showed that PM10, SO2, and NO2, CO, and atmospheric pressure （PRES） were highly correlated with PM2.5 concentration, and dew point temperature （DEWP） was strongly correlated with PM2.5 concentration, which was consistent with the preference setting of the model's automatic screening. The MSE and R2 of the transformer model are 0.009 μg·m-3 and 0.925, respectively, which decreased by 91.09% and 30.77% of MSE and increased by 38.05% and 4.65% of R2, respectively, when compared with ResNet50 and LSTM. The transformer model could capture short-term pollution changes caused by sudden changes in meteorological conditions and long-term trends with significant seasonal changes. The fitting effect of the transformer was excellent among several models, providing a novel method for long-term prediction of PM2.5 concentration. In addition, ablation experiments revealed that the increase in R2 of the transformer was relatively small after data input or manually setting preferences, with only a 2.31% and 1.51% increase, respectively, indicating that the transformer model had strong anti-interference ability for PM10 homologous data.

LULC-SegNet: Enhancing Land Use and Land Cover Semantic Segmentation with Denoising Diffusion Feature Fusion

Deep convolutional networks often encounter information bottlenecks when extracting land object features, resulting in critical geometric information loss, which impedes semantic segmentation capabilities in complex geospatial backgrounds. We developed LULC-SegNet, a semantic segmentation network for land use and land cover (LULC), which integrates features from the denoising diffusion probabilistic model (DDPM). This network enhances the clarity of the edge segmentation, detail resolution, and the visualization and accuracy of the contours by delving into the spatial details of the remote sensing images. The LULC-SegNet incorporates DDPM decoder features into the LULC segmentation task, utilizing machine learning clustering algorithms and spatial attention to extract continuous DDPM semantic features. The network addresses the potential loss of spatial details during feature extraction in convolutional neural network (CNN), and the integration of the DDPM features with the CNN feature extraction network improves the accuracy of the segmentation boundaries of the geographical features. Ablation and comparison experiments conducted on the Circum-Tarim Basin Region LULC Dataset demonstrate that the LULC-SegNet improved the LULC semantic segmentation. The LULC-SegNet excels in multiple key performance indicators compared to existing advanced semantic segmentation methods. Specifically, the network achieved remarkable scores of 80.25% in the mean intersection over union (MIOU) and 93.92% in the F1 score, surpassing current technologies. The LULC-SegNet demonstrated an IOU score of 73.67%, particularly in segmenting the small-sample river class. Our method adapts to the complex geophysical characteristics of remote sensing datasets, enhancing the performance of automatic semantic segmentation tasks for land use and land cover changes and making critical advancements.

SAITI-DCGAN: Self-Attention Based Deep Convolutional Generative Adversarial Networks for Data Augmentation of Infrared Thermal Images

Defect detection plays a crucial role in industrial production, and the implementation of this technology has significant implications for improving both product quality and processing efficiency. However, the limited availability of defect samples for training deep-learning-based object detection models within industrial processes poses challenges for model training. In this paper, we propose a novel deep convolutional generative adversarial network with self-attention mechanism for the data augmentation of infrared thermal images for the application of aluminum foil sealing. To further expand its applicability, the proposed method is designed not only to address the specific needs of aluminum foil sealing but also to serve as a robust framework that can be adapted to a wide range of industrial defect detection tasks. To be specific, the proposed approach integrates a self-attention module into the generator, adopts spectral normalization in both the generator and discriminator, and introduces a two time-scale update rule to coordinate the training process of these components. The experimental results validated the superiority of the proposed approach in terms of the synthesized image quality and diversity. The results show that our approach can capture intricate details and distinctive features of defect images of aluminum foil sealing. Furthermore, ablation experiments demonstrated that the combination of self-attention, spectral normalization, and two time-scale update rules significantly enhanced the quality of image generation, while achieving a balance between stability and training efficiency. This innovative framework marks a notable technical breakthrough in the field of industrial defect detection and image synthesis, offering broad application prospects.

A motor imagery classification model based on hybrid brain-computer interface and multitask learning of electroencephalographic and electromyographic deep features

ObjectiveExtracting deep features from participants’ bioelectric signals and constructing models are key research directions in motor imagery (MI) classification tasks. In this study, we constructed a multimodal multitask hybrid brain-computer interface net (2M-hBCINet) based on deep features of electroencephalogram (EEG) and electromyography (EMG) to effectively accomplish motor imagery classification tasks.MethodsThe model first used a variational autoencoder (VAE) network for unsupervised learning of EEG and EMG signals to extract their deep features, and subsequently applied the channel attention mechanism (CAM) to select these deep features and highlight the advantageous features and minimize the disadvantageous ones. Moreover, in this study, multitask learning (MTL) was applied to train the 2M-hBCINet model, incorporating the primary task that is the MI classification task, and auxiliary tasks including EEG reconstruction task, EMG reconstruction task, and a feature metric learning task, each with distinct loss functions to enhance the performance of each task. Finally, we designed module ablation experiments, multitask learning comparison experiments, multi-frequency band comparison experiments, and muscle fatigue experiments. Using leave-one-out cross-validation(LOOCV), the accuracy and effectiveness of each module of the 2M-hBCINet model were validated using the self-made MI-EEMG dataset and the public datasets WAY-EEG-GAL and ESEMIT.ResultsThe results indicated that compared to comparative models, the 2M-hBCINet model demonstrated good performance and achieved the best results across different frequency bands and under muscle fatigue conditions.ConclusionThe 2M-hBCINet model constructed based on EMG and EEG data innovatively in this study demonstrated excellent performance and strong generalization in the MI classification task. As an excellent end-to-end model, 2M-hBCINet can be generalized to be used in EEG-related fields such as anomaly detection and emotion analysis.

Hybrid Deep Learning Model for Cataract Diagnosis Assistance

With the population aging globally, cataracts have become one of the main causes of vision impairment. Early diagnosis and treatment of cataracts are crucial for preventing blindness. However, the use of deep learning models for assisting in the diagnosis of cataracts is limited due to reasons such as scarce data labeling, small sample size, uneven distribution, and poor generalization ability in the field. Therefore, this paper proposes a hybrid deep learning network for assisting in the diagnosis of cataract fundus images, attempting to solve the above problems and limitations. The network is based on the idea of transfer learning for feature extraction of fundus images, and introduces the Squeeze-and-Excitation (SE) module and prototype network for feature enhancement and classification, improving the model’s generalization ability for new disease samples. Finally, this paper verifies the role of each part of the model through ablation experiments, especially the significant contribution of the SE_block module and the prototype network classifier in enhancing the model’s performance. The experimental results show that the proposed model achieves excellent performance in the task of cataract fundus image recognition, with an accuracy of 0.9875, AUC value of 0.9984, and F1 score of 0.9855. The establishment of this hybrid model not only provides an effective tool for the auxiliary diagnosis of cataracts but also provides a new perspective and method for the application of deep learning in the field of ophthalmic disease recognition.

A novel knowledge distillation framework for enhancing small object detection in blurry environments with unmanned aerial vehicle-assisted images

Deep learning-based object detectors excel on mobile devices but often struggle with blurry images that are common in real-world scenarios, like unmanned aerial vehicle (UAV)-assisted images. Current models are designed for sharp images, leading to potential detection failures in blurry images. Using image deblurring before object detection is an option, but it demands significant computing power and relies heavily on the accuracy of the deblurring algorithms. Another common issue is the suitable dataset for the specific problem. To address the aforementioned issues, we develop a UAV-assisted small object detection dataset and propose a novel knowledge distillation method for object detection in blurry images in complex environments. Following this, we employ a technique known as self-supervised knowledge distillation, where we introduce a deblurring subnet module with the help of two attention modules, where both networks are trained in a fully-supervised manner. Based on the experiment results, our proposed model achieves an improvement of 4.3% accuracy in the VisDrone synthetic motion blur dataset and 4.6% in detecting objects within synthetic blurry images in our developed small object detection dataset (SOD-Dataset), as well as competitive results compared with other state-of-the-art methods. Meanwhile, ablation experiments and a visualization analysis validate the contributions of each component of the model.

An attention mechanism-based lightweight UNet for musculoskeletal ultrasound image segmentation.

Accurate musculoseletal ultrasound (MSKUS) image segmentation is crucial for diagnosis and treatment planning. Compared with traditional segmentation methods, deploying deep learning segmentation methods that balance segmentation efficiency, accuracy, and model size on edge devices has greateradvantages. This paper aims to design a MSKUS image segmentation method that has fewer parameters, lower computation complexity and higher segmentationaccuracy. In this study, an attention mechanism-based lightweight UNet (AML-UNet) is designed to segment target muscle regions in MSKUS images. To suppress the transmission of redundant feature, Channel Reconstruction and Spatial Attention Module is designed in the encoding path. In addition, considering the inherent characteristic of MSKUS image, Multiscale Aggregation Module is developed to replace the skip connection architecture of U-Net. Deep supervision is also introduced to the decoding path to refine predicted masks gradually. Our method is evaluated on two MSKUS 2D-image segmentation datasets, including 3917 MSKUS and 1534 images respectively. In the experiments, a five-fold cross-validation method is adopted in ablation experiments and comparison experiments. In addition, Wilcoxon Signed-Rank Test and Bonferroni correction are employed to validate the significance level. 0.01 was used as the statistical significance level in ourpaper. AML-UNet yielded a mIoU of 84.17% and 90.14% on two datasets, representing a 3.38% ( ) and 3.48% ( ) over the Unext model. The number of parameters and FLOPs are only 0.21M and 0.96G, which are 1/34 and 1/29 of those in comparison withUNet. Our proposed model achieved superior results with fewer parameters while maintaining segmentation efficiency and accuracy compared to othermethods.

DESRGAN: Detail-enhanced generative adversarial networks for small sample single image super-resolution

Image super-resolution involves reconstructing a blurry, low-resolution image with limited information into a clear, high-resolution image containing more detailed information. The images generated by super-resolution reconstruction can enhance the performance of downstream computer vision tasks, and hold wide application prospects in fields such as industrial fault detection, plant phenotype parameter extraction, medical imaging, and more. High-frequency components in images, such as edges and texture details, typically require more attention. However, when the training samples are limited, effectively recovering clear high-frequency details of images becomes highly challenging. Therefore, this paper proposes a single-image super-resolution method based on generative adversarial networks, named DESRGAN. Compared to existing methods, DESRGAN achieves better reconstruction of image details even with a limited number of training samples. DESRGAN introduces several key innovations: a shallow generator structure to address overfitting issues in small sample scenarios, a dual-stream feature extraction network with dilated convolutions to capture multi-scale contextual information and expand the receptive field, and an artifact loss designed to eliminate artifacts and preserve the true high-frequency details of the super-resolved images. Extensive ablation experiments and comparative studies with multiple state-of-the-art models are conducted on two small sample datasets, "Root" and "Leaves," as well as five publicly available datasets. The results demonstrate that the proposed DESRGAN achieves superior performance in small sample single image super-resolution tasks, with improvements of 1.39 dB in PSNR and 0.013 in SSIM. The generated high-resolution images exhibit clear texture and edge structures, presenting favorable subjective visual effects. Moreover, the model displays strong generalization capabilities.

Identification of the number of leaks in water supply pipes based on wavelet scattering network and Bi-LSTM model with Bayesian optimization

Water supply pipeline leaks have significant impacts on urban areas and daily life. Acoustic techniques are widely applied in its leak detection, but there is a lack of research on quantifying multiple leaks in pipelines. This study presents a novel approach using a Wavelet Scattering Network (WSN) and Bi-directional Long and Short-Term Memory (Bi-LSTM) model with Bayesian optimization to identify leak quantities accurately. The WSN autonomously extracts signal features, which are subsequently processed by the Bi-LSTM for classification. Through Bayesian optimization, the model’s hyperparameters are refined to enhance performance. The results exhibit an impressive 90.8 % classification accuracy, showcasing the model’s effectiveness in accurately identifying leak quantities. In comparison with existing models, the proposed method stands out for its superior recognition accuracy, shorter training time, and enhanced noise immunity. Ablation experiments further confirm the distinct roles and contributions of the WSN, Bayesian optimization, and Bi-LSTM components in achieving these exceptional results.

Multilevel network for large deformation image registration based on feature consistency and flow normalization.

Deformable image registration is an essential technique of medical image analysis, which plays important roles in several clinical applications. Existing deep learning-based registration methods have already achieved promising performance for the registrations with small deformations, while it is still challenging to deal with the large deformation registration due to the limits of the image intensity-similarity-based objectivefunction. To achieve the image registration with large-scale deformations, we proposed a multilevel network architecture FCNet to gradually refine the registration results based on semantic feature consistency constraint and flow normalization (FN)strategy. At each level of FCNet, the architecture is mainly composed to a FeaExtractor, a FN module, and a spatial transformation module. FeaExtractor consists of three parallel streams which are used to extract the individual features of fixed and moving images, as well as their joint features, respectively. Using these features, the initial deformation field is estimated, which passes through a FN module to refine the deformation field based on the difference map of deformation filed between two adjacent levels. This allows the FCNet to progressively improve the registration performance. Finally, a spatial transformation module is used to get the warped image based on the deformation field. Moreover, in addition to the image intensity-similarity-based objective function, a semantic-feature consistency constraint is also introduced, which can further promote the alignments by imposing the similarity between the fixed and warped image features. To validate the effectiveness of the proposed method, we compared our method with the state-of-the-art methods on three different datasets. In EMPIRE10 dataset, 20, 3, and 7 fixed and moving 3D computer tomography (CT) image pairs were used for training, validation, and testing respectively; in IXI dataset, atlas to individual image registration task was performed, with 3D MR images of 408, 58, and 115 individuals were used for training, validation, and testing respectively; in the in-house dataset, patient to atlas registration task was implemented, with the 3D MR images of 94, 3, and 15 individuals being training, validation, and testing sets,respectively. The qualitative and quantitative comparison results demonstrated that the proposed method is beneficial for handling large deformation image registration problems, with the DSC and ASSD improved by at least 1.0% and 25.9% on EMPIRE10 dataset. The ablation experiments also verified the effectiveness of the proposed feature combination strategy, feature consistency constraint, and FNmodule. Our proposed FCNet enables multiscale registration from coarse to fine, surpassing existing SOTA registration methods and effectively handling long-range spatial relationships.

Sensor attack online classification for UAVs using machine learning

Unmanned Aerial Vehicle (UAV) sensors play a vital role in maintaining flight safety and stability. However, the increasing frequency and complexity of sensor attacks have emerged as a critical threat to UAV systems. The current lack of robust multi-classification methods for detecting sensor attacks limits the effectiveness and completeness of existing defense strategies. This research addresses these challenges by leveraging machine learning (ML) techniques to classify various sensor attacks using heterogeneous sensor data and control parameters, thereby enhancing UAV system security. In this study, we design and implement multiple sensor attack scenarios targeting gyroscopes, accelerometers, barometers, and GPS. Comprehensive datasets are collected during UAV flight, integrating diverse sensor readings, flight states, and control parameters. By analyzing the characteristics of sensor attacks and their impact on position estimation and attitude control, we identify and extract key features. To optimize the classification model, we employ feature importance analysis, correlation analysis, and ablation experiments, significantly reducing data dimensionality and enhancing model training efficiency. The experimental results demonstrate the proposed ML-based multi-classification model’s superior performance, achieving a detection rate of 89.38%, significantly outperforming traditional single-attack detection methods in terms of generalization capability. Our approach efficiently handles complex multi-sensor attack scenarios. Moreover, deploying the optimized model on UAV firmware enables real-time monitoring and classification, achieving an online detection rate of 74% with a response time of approximately 0.495ms per detection. The model’s lightweight design, requiring only 48KB of storage, makes it ideal for resource-constrained UAV environments. These contributions highlight the potential of our approach to enhance real-time anomaly detection and improve UAV system resilience against diverse sensor attacks.