Robust Detection Research Articles

Synthetic DNA barcodes identify singlets in scRNA-seq datasets and evaluate doublet algorithms

Single-cell RNA sequencing (scRNA-seq) datasets contain true single cells, or singlets, in addition to cells that coalesce during the protocol, or doublets. Identifying singlets with high fidelity in scRNA-seq is necessary to avoid false negative and false positive discoveries. Although several methodologies have been proposed, they are typically tested on highly heterogeneous datasets and lack a priori knowledge of true singlets. Here, we leveraged datasets with synthetically introduced DNA barcodes for a hitherto unexplored application: to extract ground-truth singlets. We demonstrated the feasibility of our framework, "singletCode," to evaluate existing doublet detection methods across a range of contexts. We also leveraged our ground-truth singlets to train a proof-of-concept machine learning classifier, which outperformed other doublet detection algorithms. Our integrative framework can identify ground-truth singlets and enable robust doublet detection in non-barcoded datasets.

Robust Detection of Critical Events in the Context of Railway Security Based on Multimodal Sensor Data Fusion.

Effective security surveillance is crucial in the railway sector to prevent security incidents, including vandalism, trespassing, and sabotage. This paper discusses the challenges of maintaining seamless surveillance over extensive railway infrastructure, considering both technological advances and the growing risks posed by terrorist attacks. Based on previous research, this paper discusses the limitations of current surveillance methods, particularly in managing information overload and false alarms that result from integrating multiple sensor technologies. To address these issues, we propose a new fusion model that utilises Probabilistic Occupancy Maps (POMs) and Bayesian fusion techniques. The fusion model is evaluated on a comprehensive dataset comprising three use cases with a total of eight real life critical scenarios. We show that, with this model, the detection accuracy can be increased while simultaneously reducing the false alarms in railway security surveillance systems. This way, our approach aims to enhance situational awareness and reduce false alarms, thereby improving the effectiveness of railway security measures.

Development of a 45K pepper GBTS liquid-phase gene chip and its application in genome-wide association studies.

Pepper (Capsicum spp.) is a vegetable that is cultivated globally and has undergone extensive domestication, leading to a significant diversification in its agronomic traits. With the advancement of genomics in pepper and the reduction in sequencing costs, the high-throughput detection of single nucleotide polymorphisms (SNPs) and small insertions-deletions (indels) has become increasingly critical for analyzing pepper germplasms and improving breeding programs. As a result, there is a pressing need for a cost-effective, high-throughput, and versatile technique suitable for both foreground and background selection in pepper breeding. In the present study, Python-based web scraping scripts were utilized to systematically extract data from published literatures and relevant sequence databases focusing on pepper genomes. Subsequent to data extraction, SNPs and indels were meticulously identified and filtered. This process culminated in the delineation of core polymorphic sites, which were instrumental in the development of specific probes. Following this, comprehensive phenotypic and genotypic analyses were conducted on a diverse collection of 420 pepper germplasms. Concurrently, a genome-wide association study (GWAS) was conducted to elucidate the genetic determinants of helical fruit shape in peppers. In this study, a 45K pepper Genotyping-By-Target-Sequencing (GBTS) liquid-phase gene chip was developed on the GenoBaits platform. This chip is composed of 45,389 probes, of which 42,535 are derived from core polymorphic sites (CPS) in the background genetic landscape, while 2,854 are associated with foreground agronomic traits, spanning across 43 traits. The CPS probes are spaced at an average interval of 68 Kb. We have assessed the performance of this chip on 420 pepper germplasms, with successful capture of target DNA fragments by 45,387 probes. Furthermore, the probe capture ratio surpassed 70% in 410 of the 420 germplasms tested. Using this chip, we have efficiently genotyped 273 germplasms for spiciness levels and elucidated the genetic relationships among 410 pepper germplasms. Our results allowed for precise clustering of sister lines and C. chinense germplasms. In addition, through a GWAS for helical fruit shape, we identified three quantitative trait loci (QTLs): heli2.1, heli11.1, and heli11.2. Within the heli11.1 QTL, a gene encoding the tubulin alpha chain was identified, suggesting its potential role in the helical growth pattern of pepper fruits. In summary, the 45K pepper GBTS liquid-phase gene chip offers robust detection of polymorphic sites and is a promising tool for advancing research into pepper germplasm and the breeding of new pepper varieties.

Selection of Stable Reference Genes for QRT-PCR in Tree Peony 'Doulv' and Functional Analysis of PsCUC3.

(1) Background: Tree peonies display extensive cultivar diversity due to widespread hybridization, resulting in a complex genetic architecture. This complexity complicates the selection of universal reference genes across different cultivars for qRT-PCR analyses. Paeonia suffruticosa 'Doulv', notable for its unique green blooms in China, exhibits chlorosis post-flowering and features petaloid stamens and pistils. (2) Methods: Based on published literature and RNA-seq data from 'Doulv', nine candidate reference genes-ACT (Actin), TUB (β-Tubulin), UBC (Ubiquitin Conjugating Enzyme), UBQ (Ubiquitin), UPL (Ubiquitin Protein Ligase), PP2A (Protein Phosphatase 2A), PP2C (Protein Phosphatase 2C), MBF1A (Multiprotein Bridging Factor 1A), and GAPDH (Glyceraldehyde-3-Phosphate Dehydrogenase)-were selected. Their expression stability was assessed across various tissues and developmental stages of 'Doulv' flowers using qRT-PCR, with evaluations conducted via GeNorm_v3.5, NormFinder_v20, and BestKeeper_v1.0. Gene cloning and expression analyses of PsCUC3, including its subcellular localization, were performed. (3) Results: GAPDH and ACT were identified as the most stable reference genes in petaloid stamens across various developmental stages of 'Doulv', whereas UBC and MBF1A were optimal across different tissues. Notably, specific conserved amino acids in PsCUC3 from 'Doulv' diverged from those in NAM/CUC3 proteins of other species, impacting its protein structure. PsCUC3 expression analysis revealed no correlation with chlorophyll content in petaloid stamens but an association with petaloid organ development. Furthermore, PsCUC3 was predominantly localized in the nucleus. (4) Conclusions: This study comprehensively evaluated suitable reference genes using GeNorm_v3.5, NormFinder_v20, and BestKeeper_v1.0 software, establishing a robust qRT-PCR detection system for 'Doulv' peony. These results provide a solid experimental foundation for further research on 'Doulv' peony. Building on this experimental foundation, the functional analysis of the PsCUC3 gene was conducted. The findings suggest a potential association between the PsCUC3 gene and floral morphology alterations in 'Doulv', identifying PsCUC3 as crucial for understanding the molecular mechanisms influencing floral structure in tree peonies.

Robust fault detection and adaptive fixed-time fault-tolerant control for quadrotor UAVs

This note scrutinizes an adaptive fault-tolerant control (FTC) approach tailored for unmanned aerial vehicles (UAVs), addressing the critical need for both fault accommodation and disturbance suppression. Departing from traditional reliance on robust discontinuous control strategies prone to chattering and demanding precise uncertainty bounds, our FTC method ensures fixed-time stability, guaranteeing the convergence of attitude tracking errors to zero. Central to our approach is an adaptive algorithm adept at concurrently estimating unknown actuator faults and upper bounds of lumped uncertainties. Moreover, our adaptive schemes accurately estimate the upper bound of the lumped uncertainty term, encompassing model uncertainties, external disturbances, and unmodeled dynamics, thereby eliminating the need for assuming known bounds on uncertainties. Stability analysis under the developed control law is thoroughly performed using the Lyapunov stability theory. Notably, our strategy employs an extended Kalman filter (EKF) observer for state estimation and fault detection, facilitating fault detection through an adaptive threshold technique dynamically adjusted based on real-time mean and variance of the residual signal. Through comprehensive simulation and experimental validations, our proposed methodology demonstrates significant advancements in ensuring safety and reliability in UAVs.

Enhancing Alzheimer's disease diagnosis and staging: a multistage CNN framework using MRI.

This study addresses the pervasive and debilitating impact of Alzheimer's disease (AD) on individuals and society, emphasizing the crucial need for timely diagnosis. We present a multistage convolutional neural network (CNN)-based framework for AD detection and sub-classification using brain magnetic resonance imaging (MRI). After preprocessing, a 26-layer CNN model was designed to differentiate between healthy individuals and patients with dementia. After detecting dementia, the 26-layer CNN model was reutilized using the concept of transfer learning to further subclassify dementia into mild, moderate, and severe dementia. Leveraging the frozen weights of the developed CNN on correlated medical images facilitated the transfer learning process for sub-classifying dementia classes. An online AD dataset is used to verify the performance of the proposed multistage CNN-based framework. The proposed approach yielded a noteworthy accuracy of 98.24% in identifying dementia classes, whereas it achieved 99.70% accuracy in dementia subclassification. Another dataset was used to further validate the proposed framework, resulting in 100% performance. Comparative evaluations against pre-trained models and the current literature were also conducted, highlighting the usefulness and superiority of the proposed framework and presenting it as a robust and effective AD detection and subclassification method.

Object Detection in Multispectral Remote Sensing Images Based on Cross-Modal Cross-Attention.

In complex environments a single visible image is not good enough to perceive the environment, this paper proposes a novel dual-stream real-time detector designed for target detection in extreme environments such as nighttime and fog, which is able to efficiently utilise both visible and infrared images to achieve Fast All-Weatherenvironment sensing (FAWDet). Firstly, in order to allow the network to process information from different modalities simultaneously, this paper expands the state-of-the-art end-to-end detector YOLOv8, the backbone is expanded in parallel as a dual stream. Then, for purpose of avoid information loss in the process of network deepening, a cross-modal feature enhancement module is designed in this study, which enhances each modal feature by cross-modal attention mechanisms, thus effectively avoiding information loss and improving the detection capability of small targets. In addition, for the significant differences between modal features, this paper proposes a three-stage fusion strategy to optimise the feature integration through the fusion of spatial, channel and overall dimensions. It is worth mentioning that the cross-modal feature fusion module adopts an end-to-end training approach. Extensive experiments on two datasets validate that the proposed method achieves state-of-the-art performance in detecting small targets. The cross-modal real-time detector in this study not only demonstrates excellent stability and robust detection performance, but also provides a new solution for target detection techniques in extreme environments.

Robust multiple fault detection and estimation approaches based on adaptive finite‐time observer for nonlinear systems

AbstractThis article investigates robust multiple fault detection (FD) and fault estimation (FE) schemes, based on adaptive finite‐time observer (AFTO), applied to nonlinear systems with disturbances, actuator faults, and sensor faults. The AFTO is designed, and a parameter adaptive law is constructed to continuously optimize the residual based performance function to improve the AFTO robustness. The schemes are able to simultaneously detect and estimate unknown states, disturbances, and faults. The AFTO's design, finite‐time convergence and stability conditions are guaranteed in terms of linear matrix inequalities. Finally, simulation results, based on two typical practical systems, are given to illustrate the effectiveness of the proposed approaches. The simulation results also show that the designed AFTO can offer faster convergence speed, higher estimation accuracy, and stronger robustness than conventional observer and sliding mode observer.

Reinforcement Learning and Genetic Algorithm-Based Network Module for Camera-LiDAR Detection

Cameras and LiDAR sensors have been used in sensor fusion for robust object detection in autonomous driving. Object detection networks for autonomous driving are often trained again by adding or changing datasets aimed at robust performance. Repeat training is necessary to develop an efficient network module. Existing efficient network module development changes to hand design and requires much module design experience. For this, a neural architecture search was designed, but it takes much time and requires optimizing the design process. To solve this problem, we propose a two-stage optimization method for the offspring generation process in a neural architecture search based on reinforcement learning. In addition, we propose utilizing two split datasets to solve the fast convergence problem as the objective function of the genetic algorithm: source data (daytime, sunny) and target data (day/night, adversary weather). The proposed method is an efficient module generation method requiring less time than the NSGA-NET. We confirmed the performance improvement and the convergence speed reduction using the Dense dataset. Through experiments, it was proven that the proposed method generated an efficient module.

Sign Detection Model for the Deaf People using Deep Learning

Sign detection is a crucial task in various computer vision applications, including autonomous driving, robotics, and augmented reality. This paper proposes a novel sign detection model based on deep learning techniques. The proposed model leverages convolutional neural networks (CNNs) to automatically learn discriminative features from sign images, enabling robust detection performance across diverse environmental conditions. Keywords: sign detection, deep learning, convolution neural networks, computer vision, autonomous driving, robotics, augmented reality.

Establishment and Application of a Novel Genetic Detection Panel for SNPs in Mongolian Gerbils.

The Mongolian gerbil is a distinctive experimental animal in China, as its genetic qualities possess significant value in the field of medical biology research. Here, we aimed to establish an economical and efficient panel for genetic quality detection in Mongolian gerbils using single-nucleotide polymorphism (SNP) markers. To search for SNPs, we conducted whole-genome sequencing (WGS) in 40 Mongolian gerbils from outbred populations. Reliable screening criteria were established to preliminarily select SNPs with a wide genome distribution and high levels of polymorphism. Subsequently, a multiple-target regional capture detection system based on second-generation sequencing was developed for SNP genotyping. Based on the results of WGS, 219 SNPs were preliminarily selected, and they were established and optimized in a multiple-amplification system that included 206 SNP loci by genotyping three outbred populations. PopGen.32 analysis revealed that the average effective allele number, Shannon index, observed heterozygosity, expected heterozygosity, average heterozygosity, polymorphism information content, and other population genetic parameters of the Capital Medical University (CMU) gerbils were the highest, followed by those of Zhejiang gerbils and Dalian gerbils. Through scientific screening and optimization, we successfully established a novel, robust, and cost-effective genetic detection system for Mongolian gerbils by utilizing SNP markers for the first time.

A Robust Target Detection Algorithm Based on the Fusion of Frequency-Modulated Continuous Wave Radar and a Monocular Camera

Decision-level information fusion methods using radar and vision usually suffer from low target matching success rates and imprecise multi-target detection accuracy. Therefore, a robust target detection algorithm based on the fusion of frequency-modulated continuous wave (FMCW) radar and a monocular camera is proposed to address these issues in this paper. Firstly, a lane detection algorithm is used to process the image to obtain lane information. Then, two-dimensional fast Fourier transform (2D-FFT), constant false alarm rate (CFAR), and density-based spatial clustering of applications with noise (DBSCAN) are used to process the radar data. Furthermore, the YOLOv5 algorithm is used to process the image. In addition, the lane lines are utilized to filter out the interference targets from outside lanes. Finally, multi-sensor information fusion is performed for targets in the same lane. Experiments show that the balanced score of the proposed algorithm can reach 0.98, which indicates that it has low false and missed detections. Additionally, the balanced score is almost unchanged in different environments, proving that the algorithm is robust.

Enhancing Automated Brain Tumor Detection Accuracy Using Artificial Intelligence Approaches for Healthcare Environments.

Medical imaging and deep learning models are essential to the early identification and diagnosis of brain cancers, facilitating timely intervention and improving patient outcomes. This research paper investigates the integration of YOLOv5, a state-of-the-art object detection framework, with non-local neural networks (NLNNs) to improve brain tumor detection's robustness and accuracy. This study begins by curating a comprehensive dataset comprising brain MRI scans from various sources. To facilitate effective fusion, the YOLOv5 and NLNNs, K-means+, and spatial pyramid pooling fast+ (SPPF+) modules are integrated within a unified framework. The brain tumor dataset is used to refine the YOLOv5 model through the application of transfer learning techniques, adapting it specifically to the task of tumor detection. The results indicate that the combination of YOLOv5 and other modules results in enhanced detection capabilities in comparison to the utilization of YOLOv5 exclusively, proving recall rates of 86% and 83% respectively. Moreover, the research explores the interpretability aspect of the combined model. By visualizing the attention maps generated by the NLNNs module, the regions of interest associated with tumor presence are highlighted, aiding in the understanding and validation of the decision-making procedure of the methodology. Additionally, the impact of hyperparameters, such as NLNNs kernel size, fusion strategy, and training data augmentation, is investigated to optimize the performance of the combined model.

Mitigation of Adversarial Attacks in 5G Networks with a Robust Intrusion Detection System Based on Extremely Randomized Trees and Infinite Feature Selection

Intrusion Detection Systems (IDSs) are an important tool to mitigate cybersecurity threats in the ICT infrastructures. Preferable properties of the IDSs are the optimization of the attack detection accuracy and the minimization of the computing resources and time. A signification portion of IDSs presented in the research literature is based on Machine Learning (ML) and Deep Learning (DL) elements, but they may be prone to adversarial attacks, which may undermine the overall performance of the IDS algorithm. This paper proposes a novel IDS focused on the detection of cybersecurity attacks in 5G networks, which addresses in a simple but effective way two specific adversarial attacks: (1) tampering of the labeled set used to train the ML algorithm, (2) modification of the features in the training data set. The approach is based on the combination of two algorithms, which have been introduced recently in the research literature. The first algorithm is the Extremely Randomized Tree (ERT) algorithm, which enhances the capability of Decision Tree (DT) and Random Forest (RF) algorithms to perform classification in data sets, which are unbalanced and of large size as IDS data sets usually are (legitimate traffic messages are more numerous than attack related messages). The second algorithm is the recently introduced Infinite Feature Selection algorithm, which is used to optimize the choice of the hyper-parameter defined in the approach and improve the overall computing efficiency. The result of the application of the proposed approach on a recently published 5G IDS data set proves its robustness against adversarial attacks with different degrees of severity calculated as the percentage of the tampered data set samples.

A multi-weapon detection using ensembled learning

Abstract Recently, the level of criminals and terrorists using light weapons (such as knives and firearms) has increased rapidly around the world. Unfortunately, most current surveillance systems are still based mainly on human monitoring and intervention. For that reason, the requirement for a smart system for detecting different weapons becomes crucial in the field of security and computer vision. In this article, a novel technique for detecting various types of weapons has been proposed. This system is based mainly on deep learning techniques, namely, You Only Look Once, version 8 (YOLOv8), to detect a different class of light weapons. Furthermore, this study focuses on detecting two armed human poses based on ensemble learning techniques, which involve combining the outputs of different Yolov8 models to produce an accurate and robust detection system. The proposed system is evaluated on the self-created weapons dataset comprising thousands of images of different classes of weapons. The experiment results of this work show the effectiveness of ensemble learning for detecting various weapons with high accuracy, achieving 97.2% of mean average precision.

Fractional B-Spline Wavelets and U-Net Architecture for Robust and Reliable Vehicle Detection in Snowy Conditions.

This paper addresses the critical need for advanced real-time vehicle detection methodologies in Vehicle Intelligence Systems (VIS), especially in the context of using Unmanned Aerial Vehicles (UAVs) for data acquisition in severe weather conditions, such as heavy snowfall typical of the Nordic region. Traditional vehicle detection techniques, which often rely on custom-engineered features and deterministic algorithms, fall short in adapting to diverse environmental challenges, leading to a demand for more precise and sophisticated methods. The limitations of current architectures, particularly when deployed in real-time on edge devices with restricted computational capabilities, are highlighted as significant hurdles in the development of efficient vehicle detection systems. To bridge this gap, our research focuses on the formulation of an innovative approach that combines the fractional B-spline wavelet transform with a tailored U-Net architecture, operational on a Raspberry Pi 4. This method aims to enhance vehicle detection and localization by leveraging the unique attributes of the NVD dataset, which comprises drone-captured imagery under the harsh winter conditions of northern Sweden. The dataset, featuring 8450 annotated frames with 26,313 vehicles, serves as the foundation for evaluating the proposed technique. The comparative analysis of the proposed method against state-of-the-art detectors, such as YOLO and Faster RCNN, in both accuracy and efficiency on constrained devices, emphasizes the capability of our method to balance the trade-off between speed and accuracy, thereby broadening its utility across various domains.

Robust dimensionality-reduced epilepsy detection system using EEG wavelet packets and machine learning

Robust dimensionality-reduced epilepsy detection system using EEG wavelet packets and machine learning

The Performance Analysis of Graph Neural Network (GNN) and Convolutional Neural Network (CNN) Algorithms for Cyberbullying Detection in Twitter Comments

Cyberbullying incidents have surged due to the expansion of social media network and advancements in internet technology, presenting a substantial challenge in online communities. Previous studies employing Support Vector Machine (SVM) techniques have exhibited promising outcomes, achieving a superior accuracy of 71.25%. However, recognizing the dynamic nature of cyberbullying behaviors and the necessity for more robust detection methodologies, this research explores cyberbullying detection on Twitter utilizing Convolutional Neural Network (CNN) and Graph Neural Network (GNN). The selection of CNN and GNN is motivated by the deficiencies observed in prior SVM-based approaches and the capacity of neural network to capture intricate patterns in textual and network data. The GNN consistently outperforms CNN in terms of F1 score, accuracy, precision, and recall. With only 20 epochs, GNN achieves an accuracy of 80.25%, surpassing CNN's 68.43%. Through GNN optimization, its accuracy reaches 89.04% after 100 epochs, underscoring its efficacy in Twitter cyberbullying detection.

EdgeGuard: Machine Learning for Proactive Intrusion Detection on Edge Networks

Edge computing has emerged as a promising paradigm to address the challenges of latency-sensitive applications and the exponential growth of Internet of Things (IoT) devices. However, the distributed nature of edge networks introduces new security vulnerabilities, necessitating robust intrusion detection mechanisms. In this paper, we propose EdgeGuard, a machine learning based framework for proactive intrusion detection on edge networks. Leveraging convolutional neural networks (CNNs) arranged in a residual fashion, EdgeGuard effectively captures complex patterns in network traffic data, enhancing the system's ability to detect intrusions with high accuracy. We conduct experiments using the Edge-IIoTset Cyber Security Dataset, containing a diverse range of normal and attack traffic samples. The proposed method achieves promising results, as evidenced by the receiver operating characteristic area under the curve (ROCAUC) and confusion matrix analysis. EdgeGuard offers a robust solution to safeguard edge computing environments against cyber threats, contributing to the security and integrity of IoT systems in real-world deployment scenarios.

Terahertz s-SNOM Imaging of a Single Cell with Nanoscale Resolution.

Terahertz scattering scanning near-field optical microscopy is a robust spectral detection technique with a nanoscale resolution. However, there are still major challenges in investigating the heterogeneity of cell membrane components in individual cells. Here, we present a novel and comprehensive analytical approach for detecting and investigating heterogeneity in cell membrane components at the single-cell level. In comparison to the resolution of the topographical atomic force microscopy image, the spatial resolution of the terahertz near-field amplitude image is 3 times that of the former. This ultrafine resolution enables the compositional distribution in the cell membrane, such as the distribution of extracellular vesicles, to be finely characterized. Furthermore, via extraction of the near-field absorption images at specific frequencies, the visualization and compositional difference analysis of cell membrane components can be presented in detail. These findings have significant implications for the intuitive and visual analysis of cell development and disease evolutionary pathways.