Basic Network Research Articles

Blood Cell Target Detection Based on Improved YOLOv5 Algorithm

In the medical field, blood analysis is a key method used to evaluate the health status of the human body. The types and number of blood cells serve as important criteria for doctors to diagnose and treat diseases. In view of the problems regarding difficult classification and low efficiency in blood cell detection, this paper proposes an improved YOLOv5-BS blood cell target detection algorithm. The purpose of the improvement is to enhance the real-time performance and accuracy of blood cell type recognition. The algorithm is based on YOLOv5s as the basic network, incorporating the advantages of both CNN and Transformer architectures. First, the BotNet backbone network is incorporated. Then the YOLOv5 head architecture is replaced with the Decoupled Head structure. Finally, a new loss function SIoU is used to improve the accuracy and efficiency of the model. To detect the feasibility of the algorithm, a comparative experiment was conducted. The experiment shows that the improved algorithm has an accuracy of 92.8% on the test set, an average precision of 83.3%, and a recall rate of 99%. Compared with YOLOv8s and PP-YOLO, the average precision is increased by 3.9% and 1%, and the recall rate is increased by 3% and 2%. This algorithm effectively improves the efficiency and accuracy of blood cell detection and effectively improves the problem of blood cell detection.

Triplet Network for One-Shot Raman Spectrum Recognition.

Raman spectroscopy is widely used for material detection due to its specificity, but its application to spectral recognition often faces limitations due to insufficient training data, unlike fields such as image recognition. Traditional machine learning or basic neural networks are commonly used, but they have limited ability to achieve high precision. We have proposed a novel approach that combines the Triplet network (TN) and K-nearest neighbor (KNN) techniques to address this issue. TN maps the Raman spectral sequences to a 128-dimensional Euclidean space to extract features, enabling the features in the new space to more accurately represent the similarities or differences between spectra, and then utilizes the KNN algorithm to perform classification tasks in this feature space. Our method exhibits superior performance in recognizing unknown Raman spectra with minimal training samples per class. We employed a handheld Raman spectrometer with an excitation wavelength of 785 nm to collect the Raman spectra of 36 samples, including 28 safe materials and eight hazardous materials. Using only one spectrum as a support set for each category, the hazardous samples were successfully distinguished from the safe samples with an accuracy of 99.6%. Additionally, our model offers adaptability without requiring exhaustive retraining when adding new prediction classes. In situations with high background fluorescence, the TN performs better in measuring the distance between spectra of the same class than traditional distance measurement methods.

Optimization of Interference Identification for Wireless Communication Based on Global Convolutional Neural Network

With the widespread adoption of smartphones and the rise of 5G technology, wireless networks fully show their bright future. In order to study the interference factors and optimization methods in the process of wireless communication, this paper studies the wireless communication interference recognition algorithm and power allocation algorithm based on global convolutional neural network, proposes a wireless communication interference recognition system model based on global convolutional neural network, and trains the system. The specific conclusions are as follows: (1) The basic modulation recognition neural network algorithm is constructed based on deep learning theory; (2) As the number of D2D users increases, the time used by the WMMSE algorithm and the equal power allocation algorithm in the traditional method increases. (3) The QL algorithm-based on greedy- is compared with the system-based QL algorithm proposed in this paper. On 200-time nodes, the SINK value of the QL algorithm based on -greedy is increased from 2.6 to 3.3, and the convergence speed is also greatly improved. The system proposed in this paper has a faster learning speed, can accelerate the learning of strategies, and can save a lot of computing resources, and effectively reduce the complexity of operations.

Use of Image Recognition and Machine Learning for the Automatic and Objective Evaluation of Standstill Marks on Rolling Bearings

One main research area of the Competence Centre for Tribology is so-called standstill marks (SSMs) at roller bearings that occur if the bearing is exposed to vibrations or performs just micromovements. SSMs obtained from experiments are usually photographed, evaluated and manually categorized into six classes. An internal project has now investigated the extent to which this evaluation can be automated and objectified. Images of standstill marks were classified using convolutional neural networks that were implemented with the deep learning library Pytorch. With basic convolutional neural networks, an accuracy of 70.19% for the classification of all six classes and 83.65% for the classification of pairwise classes was achieved. Classification accuracies were improved by image augmentation and transfer learning with pre-trained convolutional neural networks. Overall, an accuracy of 83.65% for the classification of all six standstill mark classes and 91.35% for the classification of pairwise classes was achieved. Since 16 individual marks are generated per test run in a typical quasi standstill test (QSST) of the CCT and the deviation in the prediction of the classification is a maximum of one school grade, the accuracy achieved is already sufficient to carry out a reliable and objective evaluation of the markings.

NEW SANTA CATARINA HIGH SCHOOL: CHALLENGES EXPERIENCED BY TEACHERS AND MANAGERS

This research aims to analyze the challenges faced by teachers and administrators of the state basic education network of Santa Catarina in relation to the implementation of the New High School. The methodology used was qualitative and exploratory. The study included nine teachers and three school administrators from the state of Santa Catarina who participated in “pilot schools” of this implementation. A semi-structured interview was used to collect data. The results shown are challenging, among which stand out the lack of dialogue between educational institutions, the lack of school structure and basic investments for the development of full-time education, the lack of teaching materials and collective planning in the areas of knowledge, as well as continuing education processes. These results can contribute to stimulating active and democratic participation of the school community in the decision-making processes, reducing the distance between the discourses and the reality experienced by education professionals.

Adaptive Kernel Convolutional Stereo Matching Recurrent Network.

For binocular stereo matching techniques, the most advanced method currently is using an iterative structure based on GRUs. Methods in this class have shown high performance on both high-resolution images and standard benchmarks. However, simply replacing cost aggregation with a GRU iterative method leads to the original cost volume for disparity calculation lacking non-local geometric and contextual information. Based on this, this paper proposes a new GRU iteration-based adaptive kernel convolution deep recurrent network architecture for stereo matching. This paper proposes a kernel convolution-based adaptive multi-scale pyramid pooling (KAP) module that fully considers the spatial correlation between pixels and adds new matching attention (MAR) to refine the matching cost volume before inputting it into the iterative network for iterative updates, enhancing the pixel-level representation ability of the image and improving the overall generalization ability of the network. At present, the AKC-Stereo network proposed in this paper has a higher improvement than the basic network. On the Sceneflow dataset, the EPE of AKC-Stereo reaches 0.45, which is 0.02 higher than the basic network. On the KITTI 2015 dataset, the AKC-Stereo network outperforms the base network by 5.6% on the D1-all metric.

Adding Noise to Super-Resolution Training Set: Method to Denoise Super Resolution for Structure from Motion Preprocessing

The resolution and noise levels of input images directly affect the three-dimensional (3D) structure-from-motion (SfM) reconstruction performance. Conventional super-resolution (SR) methods focus too little on denoising, and latent image noise becomes worse when resolution is improved. This study proposes two SR denoising training algorithms to simultaneously improve resolution and noise: add-noise-before-downsampling and downsample-before-adding-noise. These portable methods preprocess low-resolution training images using real-world noise samples instead of altering the basic neural network. Hence, they concurrently improve resolution while reducing noise for an overall cleaner SfM performance. We applied these methods to the existing SR network: super-resolution convolutional neural network, enhanced deep residual super-resolution, residual channel attention network, and efficient super-resolution transformer, comparing their performances with those of conventional methods. Impressive peak signal-to-noise and structural similarity improvements of 0.12 dB and 0.56 were achieved on the noisy images of Smartphone Image Denoising Dataset, respectively, without altering the network structure. The proposed methods caused a very small loss (<0.01 dB) on clean images. Moreover, using the proposed SR algorithm makes the 3D SfM reconstruction more complete. Upon applying the methods to non-preprocessed and conventionally preprocessed models, the mean projection error was reduced by a maximum of 27% and 4%, respectively, and the number of 3D densified points was improved by 310% and 7%, respectively.

Resistance distances in stretched Cantor product networks

In this paper, we consider resistance distances in stretched Cantor product networks, a family of non-self-similar networks. By constructing the networks in a iterated way, we give an approach to encode every node in their vertex set. And then we simplify the complex resistor networks by induction on the basic network pattern. Using classical results of circuit theory, we obtain the exact formulae of the resistance distances of some pairs of nodes in stretched Cantor product networks.

A self-growth convolution network for thermal and mechanical fault detection with very limited engine data

Severe faults occur infrequently but are critical for the prognostics and health management (PHM) of power machinery. Due to the scarcity of fault data, diagnostic models are always facing a very limited data problem. Basic convolutional neural networks require a large number of samples to train, and widely used data augmentation methods are influenced by data quality, which can exacerbate overfitting. To address this issue, a self-growth convolution network (SGNet) is proposed to make the deep learning process a self-growing scheme in both depth and width dimensions. The direct similarity measurement is utilized to supervise the depth-growth in the layer-by-layer training process. The feature redundancy metric is employed to control the width expansion. The self-growth scheme is proposed to disrupt the coadaptation between layers and that between kernels in order to mitigate the overfitting issue of small-sample cases. The SGNet is verified and implemented in the PHM of a heavy-duty diesel engine. It exhibits remarkable diagnostic capabilities in extremely sample-limited scenarios. With only three training samples per faulty type, the recognition rates of SGNet for the misfire fault and the gear tooth fracture fault are 88.44% and 98.11%, respectively. Further, the feature contrast, the information transmission, the noise resistance, and the frequency domain activation heat of SGNet are discussed by the ablation experiment in detail. The results indicate a novel path to solve the data-limitation problem in the PHM of important power machinery.

Attentional network deficits in patients with migraine: behavioral and electrophysiological evidence

BackgroundPatients with migraine often experience not only headache pain but also cognitive dysfunction, particularly in attention, which is frequently overlooked in both diagnosis and treatment. The influence of these attentional deficits on the pain-related clinical characteristics of migraine remains poorly understood, and clarifying this relationship could improve care strategies.MethodsThis study included 52 patients with migraine and 34 healthy controls. We employed the Attentional Network Test for Interactions and Vigilance–Executive and Arousal Components paradigm, combined with electroencephalography, to assess attentional deficits in patients with migraine, with an emphasis on phasic alerting, orienting, executive control, executive vigilance, and arousal vigilance. An extreme gradient boosting binary classifier was trained on features showing group differences to distinguish patients with migraine from healthy controls. Moreover, an extreme gradient boosting regression model was developed to predict clinical characteristics of patients with migraine using their attentional deficit features.ResultsFor general performance, patients with migraine presented a larger inverse efficiency score, a higher prestimulus beta-band power spectral density and a lower gamma-band event-related synchronization at Cz electrode, and stronger high alpha-band activity at the primary visual cortex, compared to healthy controls. Although no behavior differences in three basic attentional networks were found, patients showed magnified N1 amplitude and prolonged latency of P2 for phasic alerting-trials as well as an increased orienting evoked-P1 amplitude. For vigilance function, improvements in the hit rate of executive vigilance-trials were exhibited in controls but not in patients. Besides, patients with migraine exhibited longer reaction time as well as larger variability in arousal vigilance-trials than controls. The binary classifier developed by such attentional deficit features achieved an F1 score of 0.762 and an accuracy of 0.779 in distinguishing patients with migraine from healthy controls. Crucially, the predicted value available from the regression model involving attentional deficit features significantly correlated with the real value for the frequency of headache.ConclusionsPatients with migraine demonstrated significant attentional deficits, which can be used to differentiate migraine patients from healthy populations and to predict clinical characteristics. These findings highlight the need to address cognitive dysfunction, particularly attentional deficits, in the clinical management of migraine.

Efficient Brain Cancer Classification Based on Advanced CNN and Transformer Technologies

Abstract. This study addresses the challenge faced by primary medical staff in accurately distinguishing different types of brain cancer by comparing the training accuracy and iteration speed of various classification models. Specifically, thesis evaluated a basic Convolutional Neural Network (CNN) alongside two transfer learning models: EfficientNetV2 and Vision Transformer. The analysis is based on publicly available data from Kaggle, which provided a robust dataset for evaluation. The results indicate that both EfficientNetV2 and Vision Transformer models achieve significantly higher iteration convergence speeds and classification accuracy compared to the basic CNN model trained solely on the database. These advanced models demonstrate a capacity for efficient and rapid brain cancer differentiation, making them suitable for settings with limited computing resources. The study highlights that the transfer learning models can effectively support outpatient doctors by providing fast and accurate etiological screening without requiring extensive hardware. Overall, the findings suggest that EfficientNetV2 and Vision Transformer models offer a practical solution for improving brain cancer classification in primary healthcare environments, where rapid and precise diagnosis is crucial. This research underscores the potential of transfer learning techniques to enhance diagnostic capabilities while accommodating resource constraints.

Maintaining Cyber Resilience in the Reconfigurable Networks with Immunization and Improved Network Game Methods.

The paper proposes a technique for protecting reconfigurable networks that implements topology rebuilding, which combines immunization and network gaming methods, as a solution for maintaining cyber resilience. Immunization presumes an adaptive set of protective reconfigurations destined to ensure the functioning of a network. It is a protective reconfiguration aimed to preserve/increase the functional quality of the system. Network nodes and edges are adaptively reorganized to counteract an invasion. This is a functional component of cyber resilience. It can be implemented as a global strategy, using knowledge of the whole network structure, or a local strategy that only works with a certain part of a network. A formal description of global and local immune strategies based on hierarchical and peer-to-peer network topologies is presented. A network game is a kind of the well-defined game model in which each situation generates a specific network, and the payoff function is calculated based on the constructed networks. A network game is proposed for analyzing a network topology. This model allows quickly identifying nodes that require disconnection or replacement when a cyber attack occurs, and understanding which network sectors might be affected by an attack. The gaming method keeps the network topology resistant to unnecessary connections. This is a structural component of cyber resilience. The basic network game method has been improved by using the criterion of maximum possible path length to reduce the number of reconfigurations. Network optimization works together with immunization to preserve the structural integrity of the network. In an experimental study, the proposed method demonstrated its effectiveness in maintaining system quality within given functional limits and reducing the cost of system protective restructuring.

How Wetting and Drainage Cycles and Wetting Angle Affect Capillary Air Trapping and Hydraulic Conductivity: A Pore Network Modeling of Experiments on Sand

Entrapped air in porous media can significantly affect water flow but simulations of air entrapment are still challenging. We developed a pore-network model using quasi-static algorithms to simulate air entrapment during spontaneous wetting and subsequent drainage processes. The model, implemented in OpenPNM, was tailored to replicate an experiment conducted on a medium-sized unconsolidated sand sample. We started building the model with three types of relatively small networks formed by 54,000 pore bodies which we used to calibrate basic network topological parameters by fitting the model to the water retention curve and the saturated hydraulic conductivity of the sand sample. Using these parameters, along with X-ray image data (µCT), a larger network formed by over 250,000 pore bodies was introduced in the form of stacked sub-networks where topological parameters were scaled along the z-axis. We investigated the impact of two different contact angles on air entrapment. For a contact angle of 0, the model showed good agreement with the experimental data, accurately predicting the amount of entrapped air and the saturated hydraulic conductivity. On the contrary, for a contact angle of π/4, the model provided reasonable accuracy for saturated hydraulic conductivity but overestimated the amount of entrapped air. Overall, this approach demonstrated that a reasonable match between simulated and experimental data can be achieved with minimal computational costs.

Neural Approach to Coordinate Transformation for LiDAR–Camera Data Fusion in Coastal Observation

The paper presents research related to coastal observation using a camera and LiDAR (Light Detection and Ranging) mounted on an unmanned surface vehicle (USV). Fusion of data from these two sensors can provide wider and more accurate information about shore features, utilizing the synergy effect and combining the advantages of both systems. Fusion is used in autonomous cars and robots, despite many challenges related to spatiotemporal alignment or sensor calibration. Measurements from various sensors with different timestamps have to be aligned, and the measurement systems need to be calibrated to avoid errors related to offsets. When using data from unstable, moving platforms, such as surface vehicles, it is more difficult to match sensors in time and space, and thus, data acquired from different devices will be subject to some misalignment. In this article, we try to overcome these problems by proposing the use of a point matching algorithm for coordinate transformation for data from both systems. The essence of the paper is to verify algorithms based on selected basic neural networks, namely the multilayer perceptron (MLP), the radial basis function network (RBF), and the general regression neural network (GRNN) for the alignment process. They are tested with real recorded data from the USV and verified against numerical methods commonly used for coordinate transformation. The results show that the proposed approach can be an effective solution as an alternative to numerical calculations, due to process improvement. The image data can provide information for identifying characteristic objects, and the obtained accuracies for platform dynamics in the water environment are satisfactory (root mean square error—RMSE—smaller than 1 m in many cases). The networks provided outstanding results for the training set; however, they did not perform as well as expected, in terms of the generalization capability of the model. This leads to the conclusion that processing algorithms cannot overcome the limitations of matching point accuracy. Further research will extend the approach to include information on the position and direction of the vessel.

Unsupervised Anomaly Detection via Normal Feature-Enhanced Reverse Teacher–Student Distillation

In modern industrial production, unsupervised anomaly detection methods have gained significant attention due to their ability to address the challenge posed by the scarcity of labeled anomaly samples. Among them, unsupervised anomaly detection methods based on reverse distillation (RD) have become a mainstream choice, which has attracted extensive research due to their excellent anomaly detection performance. However, there is a problem of “feature leakage” in the RD model, which may lead to non-anomalous regions being incorrectly identified as defects. To solve this problem, we propose a Normal Feature-Enhanced Reverse teacher–student Distillation (NFERD) method. Specifically, we designed and incorporated a normal feature bank (NFB) module into the basic RD network. This module stores normal features extracted by the teacher model, assisting the student model in learning normal features more efficiently, thereby addressing the problem of “feature leakage”. In addition, to effectively fuse the feature maps extracted by the student model with the feature maps in NFBs, we designed a Hybrid Attention Fusion Module (HAFM), which ensures the preservation of key information during the feature fusion process by the parallel processing of spatial and channel attention mechanisms. Through experimental verification on two publicly available datasets, i.e., MVTec and KSDD, our method outperformed the existing mainstream methods in both image-level and pixel-level anomaly detection. Specifically, we achieved an average I-AUROC score of 99.32% on MVTec and a 98.75% P-AUROC on the KSDD, showing clearer segmentation results, especially in complex scenarios. Furthermore, our method surpassed the second-best method by over 1.4% PRO on MVTec, demonstrating its effectiveness.

Optimizing Scorpion Toxin Processing through Artificial Intelligence.

Scorpion toxins are relatively short cyclic peptides (<150 amino acids) that can disrupt the opening/closing mechanisms in cell ion channels. These peptides are widely studied for several reasons including their use in drug discovery. Although improvements in RNAseq have greatly expedited the discovery of new scorpion toxins, their annotation remains challenging, mainly due to their small size. Here, we present a new pipeline to annotate toxins from scorpion transcriptomes using a neural network approach. This pipeline implements basic neural networks to sort amino acid sequences to find those that are likely toxins and thereafter predict the type of toxin represented by the sequence. We anticipate that this pipeline will accelerate the classification of scorpion toxins in forthcoming scorpion genome sequencing projects and potentially serve a useful role in identifying targets for drug development.

Research on the Robustness of Command and Control Networks under Cascading Failures

The current analysis of cascading failures in command and control networks pays little attention to their roles and mechanisms, resulting in challenges in quantifying survivability evaluation metrics and limiting practical application. To address these issues, this paper designs a command and control network model with a recovery strategy to improve the scientific evaluation of critical nodes and enhance the reliability of subsequent cascading failure simulations. Two capacity parameters are introduced to analyze the nonlinear behavior between network node load and capacity, and an optimal recovery strategy is proposed. This strategy prioritizes the recovery of critical nodes, thereby minimizing the overall probability of network failure. Simulations were conducted under both random failure and deliberate attack scenarios, comparing the proposed strategy with random recovery and betweenness-priority recovery strategies to identify the optimal recovery approach. The experiments showed that the optimal recovery strategy significantly enhanced the network’s survivability and recovery efficiency, allowing for the restoration of basic network functions in the shortest possible time and reducing the impact of cascading failures. By integrating the operability and uncertainty of real-world command and control networks, this method improved the network’s recovery capability and overall stability in the face of cascading failures through scientific evaluation and strategy optimization.

Environment by environment interactions (ExE) differ across genetic backgrounds (ExExG).

While the terms "gene-by-gene interaction" (GxG) and "gene-by-environment interaction" (GxE) are widely recognized in the fields of quantitative and evolutionary genetics, "environment-byenvironment interaction" (ExE) is a term used less often. In this study, we find that environmentby-environment interactions are a meaningful driver of phenotypes, and moreover, that they differ across different genotypes (suggestive of ExExG). To support this conclusion, we analyzed a large dataset of roughly 1,000 mutant yeast strains with varying degrees of resistance to different antifungal drugs. Our findings reveal that the effectiveness of a drug combination, relative to single drugs, often differs across drug resistant mutants. Remarkably, even mutants that differ by only a single nucleotide change can have dramatically different drug × drug (ExE) interactions. We also introduce a new framework that more accurately predicts the direction and magnitude of ExE interactions for some mutants. Understanding how ExE interactions change across genotypes (ExExG) is crucial not only for modeling the evolution of pathogenic microbes, but also for enhancing our knowledge of the underlying cell biology and the sources of phenotypic variance within populations. While the significance of ExExG interactions has been overlooked in evolutionary and population genetics, these fields and others stand to benefit from understanding how these interactions shape the complex behavior of living systems.

A Deep Learning Model for Classifying Black Sigatoka Disease in Banana Leaves Based on Infection Stages

Objective: This research study aims to develop an efficient deep-learning model to detect and classify stages of Black Sigatoka disease in banana plants. Methods: In this study, deep learning techniques, specifically the basic Convolutional Neural Network (CNN) and VGG16 models, were used to address the challenge of identifying Black Sigatoka disease in banana leaves early on. The tests were conducted on a dataset containing labelled images of banana leaves, assessing their effectiveness based on criteria such as accuracy, precision, recall, and F1-score after adjusting hyperparameters for optimal outcomes. Findings: The results of the trials revealed that the basic CNN model attained a training accuracy of 96% and a validation accuracy of 89%, surpassing the performance of the VGG16 model. The VGG16 model, on the other hand, had a training accuracy of 92% and a validation accuracy of 89%. Across precision, recall, and F1 score measurements, the basic CNN model consistently outperformed the VGG16 model, with scores averaging 0.90 for all three metrics compared to VGG16's precision of 0.80, recall of 0.75, and F1 score of 0.75. The CNN model demonstrated its efficiency by stopping training at 26 epochs, whereas VGG16 completed training in 21 epochs. This demonstrates its effectiveness in detecting Black Sigatoka while utilising minimal resources. Novelty: A significant component of this study is its emphasis on identifying the stages of Black Sigatoka disease, which is commonly overlooked in research. By studying disease progression, this study provides insights for early intervention and disease management, aiding efforts to lessen the impact of Black Sigatoka on banana farming. Keywords: Black Sigatoka, Deep Learning, Disease Stages, Convolutional Neural Network, Classification, Identification

Tumor Detection in MRI Data using Deep Learning Techniques for Image Classification and Semantic Segmentation

This study aims to apply image classification and semantic segmentation methods using deep learning for accurate and efficient tumor detection in brain MRI data. It looks at ways to get over issues including interpretability of the model and a lack of annotated data. A pre-trained VGG19 model and a basic convolutional neural network (CNN) were trained using a public dataset of 273 brain MRI images for the purpose of image classification. Techniques for regularization, transfer learning, and data augmentation were used. A U-Net architecture was trained using manually created masks from the same dataset for semantic segmentation, with several training setups investigated. With a mean test accuracy of 77.19%, the pre-trained VGG19 model surpassed the basic CNN in picture categorization. The U-Net model performed better for semantic segmentation after training on the entire dataset for two epochs as opposed to training on a subset for ten epochs. The work acknowledges issues including data scarcity and the requirement for model interpretability and scalability, which call for additional research, but also underlines the potential of deep learning for brain tumor identification.