Search

Marine mammals are an important part of marine ecosystems, and human intervention seriously threatens their living environments. Few studies exist on the marine mammal call recognition task, and the accuracy of current research needs to improve. In this paper, a novel MG-ResFormer two-channel fusion network architecture is proposed, which can extract local features and global timing information from sound signals almost perfectly. Second, in the input stage of the model, we propose an improved acoustic feature energy fingerprint, which is different from the traditional single feature approach. This feature also contains frequency, energy, time sequence and other speech information and has a strong identity. Additionally, to achieve more reliable accuracy in the multiclass call recognition task, we propose a multigranular joint layer to capture the family and genus relationships between classes. In the experimental section, the proposed method is compared with the existing feature extraction methods and recognition methods. In addition, this paper also compares with the latest research, and the proposed method is the most advanced algorithm thus far. Ultimately, our proposed method achieves an accuracy of 99.39% in the marine mammal call recognition task. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) paradigm and the advanced wireless technologies of 5G and beyond are expected to enable diverse applications such as autonomous driving, industrial automation, and smart cities. These applications bring together a vast and diverse IoT device population that occupy radio frequency spectrum. Such a large number of wireless devices expose previously unheard-of threat surfaces in addition to the bandwidth shortage and throughput issues. Device identification is crucial in such scenarios not only to authenticate and authorize nodes, but also to employ different network services. One of the promising solutions for device identification is the use of radio frequency (RF) fingerprinting. Recently, wireless device identification using RF fingerprinting along with machine learning and deep learning technologies showed outstanding results in the recent contemporary domains. This paper presents a systematic literature review of RF fingerprinting identification of wireless devices by presenting the results as a graphical and tabular representation of statistical data obtained. Only experimental research papers were considered of over 130 journals and international conference papers that have been classified and evaluated from the year 2010 till date. This survey focuses on exploring the commonly used RF fingerprinting approaches, feature extraction and filtration techniques, and classification algorithm used in the device identification. Finally, open issues and challenges along with future directions have presented which were discovered during the process of analyzing the literature. [ABSTRACT FROM AUTHOR]

During the manufacturing process of printed circuit boards (PCBs), quality defects can occur, which can affect the performance and reliability of PCBs. Existing deep learning-based PCB defect detection methods are difficult to simultaneously achieve the goals of high detection accuracy, fast detection speed, and small number of parameters. Therefore, this paper proposes a PCB defect detection algorithm based on CDI-YOLO. Firstly, the coordinate attention mechanism (CA) is introduced to improve the backbone and neck network of YOLOv7-tiny, enhance the feature extraction capability of the model, and thus improve the accuracy of model detection. Secondly, DSConv is used to replace part of the common convolution in YOLOv7-tiny to achieve lower computing costs and faster detection speed. Finally, Inner-CIoU is used as the bounding box regression loss function of CDI-YOLO to speed up the bounding box regression process. The experimental results show that the method achieves 98.3% mAP on the PCB defect dataset, the detection speed is 128 frames per second (FPS), the parameters is 5.8 M, and the giga floating-point operations per second (GFLOPs) is 12.6 G. Compared with the existing methods, the comprehensive performance of this method has advantages. [ABSTRACT FROM AUTHOR]

Recently, biometric recognition has become a significant field of research. The concept of cancelable biometrics (CB) has been introduced to address security concerns related to the handling of sensitive data. In this paper, we address unconstrained face verification by proposing a deep cancelable framework called BiometricNet+ that employs random projections (RP) to conceal face images and compressive sensing (CS) to reconstruct measurements in the original domain. Our lightweight design enforces the properties of unlinkability, revocability, and non-invertibility of the templates while preserving face recognition accuracy. We compare facial features by learning a regularized metric: at training time, we jointly learn facial features and the metric such that matching and non-matching pairs are mapped onto latent target distributions; then, for biometric verification, features are randomly projected via random matrices changed at every enrollment and query and reconstructed before the latent space mapping is computed. We assess the face recognition accuracy of our framework on challenging datasets such as LFW, CALFW, CPLFW, AgeDB, YTF, CFP, and RFW, showing notable improvements over state-of-the-art techniques while meeting the criteria for secure cancelable template design. Since our method requires no fine-tuning of the learned features, it can be applied to pre-trained networks to increase sensitive data protection. [ABSTRACT FROM AUTHOR]

With the rapid development of computer vision and deep learning technology, person re-identification(ReID) has attracted widespread attention as an important research area. Most current ReID methods primarily focus on short-term re-identification. In the scenario of pedestrian clothing changes, traditional ReID methods face some challenges due to significant changes in pedestrian appearance. Therefore, this paper proposes a clothes-changing person re-identification(CC-ReID) method, namely SViT-ReID, based on a Vision Transformer and incorporating semantic information. This method integrates semantic segmentation maps to more accurately extract features and representations of pedestrian instances in complex scenes, enabling the model to learn some clues unrelated to clothing. Specifically, we extract clothing-unrelated features (such as the face, arms, legs, and feet) from pedestrian parsing tasks' obtained features. These features are then fused with global features to emphasize the importance of these body features. In addition, the complete semantic features derived from pedestrian parsing are fused with global features. These fused features undergo shuffle and grouping operations to generate local features, which are computed in parallel with global features, thereby enhancing the model's robustness and accuracy. Experimental evaluations on two real-world benchmarks show the proposed SViT-ReID achieves state-of-the-art performance. Extensive ablation studies and visualizations illustrate the effectiveness of our method. Our method achieves a Top-1 accuracy of 55.2% and 43.4% on the PRCC and LTCC datasets, respectively. [ABSTRACT FROM AUTHOR]

Secure user authentication has grown importance in today’s modern culture. It is significant to authenticate the user identity in numerous consumer applications particularly financial transactions. Traditional authentication methods rely on easy-to-guess passwords, PIN numbers, or tokens with several security flaws, such as those printed on the back of credit cards for PIN numbers. As an alternative to current systems, biometric authentication techniques based on physical and behavioral characteristics have been proposed. Multibiometric systems, which combine several biometrics, are developed as a result of the difficulties that single-biometric authentication systems encountered in real-world applications including lack of precision and noisy data. The proposed system provides better performance and greater accuracy compared with other authentication techniques. The majority of them is inconvenient and demand complicated user interactions. This paper proposes Enhanced Elman Spike Neural Network along Glowworm Swarm Optimization (EESNN-GSO-AMT) for Multiple Transaction Authentication. The images are collected via SDUMLA-HMTalong CASIA V5 dataset. The pictures are provided to pre-processing to enhance the images quality utilizing Learnable Edge Collaborative Filter (LECF). The preprocessed images are fed to feature extraction using Adaptive and concise empirical wavelet transform (ACEWT) and the features are extracted such as entropy, homogeneity, energy and contrast. The extracting features are provided to EESNN classifier to categorize authorized or unauthorized persons. In general, the EESNN classifier does not express adapting optimization methods to determine ideal parameters to ensure accurately. Therefore, it is proposed to utilize the Glowworm Swarm Optimization to enhanceEESNN, which accurately categorizes the authorized and unauthorized person. The efficiency of the proposed approach is assessed usingsome metrics. The proposed EESNN-GSO-AMT method attains higher accuracy 20.54%, 21.76% and 23.89%; greater sensitivity 20.12% 20.34% and 21.43%; higher precision 23.34%, 22.68% and 24.34% are analyzed to the existing methods, like Optimal feature level fusion for safe human authentication in multimodal biometric scheme (OptGWO-AMT-FV), Joint attention network for finger vein authentication (JAnet-AMT-FV), Finger Vein Recognition Utilizing Deep Learning Technique (DCNN-AMT-FV) respectively. [ABSTRACT FROM AUTHOR]

The computing device deployment explosion experienced in recent years, motivated by the advances of technologies such as Internet-of-Things (IoT) and 5G, has led to a global scenario with increasing cybersecurity risks and threats. Among them, device spoofing and impersonation cyberattacks stand out due to their impact and, usually, low complexity required to be launched. To solve this issue, several solutions have emerged to identify device models and types based on the combination of behavioral fingerprinting and Machine/Deep Learning (ML/DL) techniques. However, these solutions are not appropriate for scenarios where data privacy and protection are a must, as they require data centralization for processing. In this context, newer approaches such as Federated Learning (FL) have not been fully explored yet, especially when malicious clients are present in the scenario setup. The present work analyzes and compares the device model identification performance of a centralized DL model with an FL one while using execution time-based events. For experimental purposes, a dataset containing execution-time features of 55 Raspberry Pis belonging to four different models has been collected and published. Using this dataset, the proposed solution achieved 0.9999 accuracy in both setups, centralized and federated, showing no performance decrease while preserving data privacy. Later, the impact of a label-flipping attack during the federated model training is evaluated using several aggregation mechanisms as countermeasures. Zeno and coordinate-wise median aggregation show the best performance, although their performance greatly degrades when the percentage of fully malicious clients (all training samples poisoned) grows over 50%. [ABSTRACT FROM AUTHOR]

Conventional machine learning (ML) and deep learning (DL) play a key role in the selectivity prediction of kinase inhibitors. A number of models based on available datasets can be used to predict the kinase profile of compounds, but there is still controversy about the advantages and disadvantages of ML and DL for such tasks. In this study, we constructed a comprehensive benchmark dataset of kinase inhibitors, involving in 141,086 unique compounds and 216,823 well-defined bioassay data points for 354 kinases. We then systematically compared the performance of 12 ML and DL methods on the kinase profiling prediction task. Extensive experimental results reveal that (1) Descriptor-based ML models generally slightly outperform fingerprint-based ML models in terms of predictive performance. RF as an ensemble learning approach displays the overall best predictive performance. (2) Single-task graph-based DL models are generally inferior to conventional descriptor- and fingerprint-based ML models, however, the corresponding multi-task models generally improves the average accuracy of kinase profile prediction. For example, the multi-task FP-GNN model outperforms the conventional descriptor- and fingerprint-based ML models with an average AUC of 0.807. (3) Fusion models based on voting and stacking methods can further improve the performance of the kinase profiling prediction task, specifically, RF::AtomPairs + FP2 + RDKitDes fusion model performs best with the highest average AUC value of 0.825 on the test sets. These findings provide useful information for guiding choices of the ML and DL methods for the kinase profiling prediction tasks. Finally, an online platform called KIPP (https://kipp.idruglab.cn) and python software are developed based on the best models to support the kinase profiling prediction, as well as various kinase inhibitor identification tasks including virtual screening, compound repositioning and target fishing. [ABSTRACT FROM AUTHOR]