Search

Pattern recognition systems always misclassify anomalies, which can be dangerous for uninformed users. Therefore, anomalies must be filtered out from each classification. The main challenge for the anomaly filter design is the huge number of possible anomaly samples compared with the number of samples in the training set. Tailoring the filter for the given classifier is just the first step in this reduction. Paper tests the hypothesis that the filter trained in avoiding "near" anomalies will also refuse the "far" anomalies, and the anomaly detector is then just a classifier distinguishing between "far real" and "near anomaly" samples. As a "far real" samples generator was used, a Generative Adversarial Network (GAN) fake generator that transforms normally distributed random seeds into fakes similar to the training samples. The paper proves the assumption that seeds unused in fake training will generate anomalies. These seeds are distinguished according to their Chebyshev norms. While the fakes have seeds within the hypersphere with a given radius, the near anomalies have seeds within the sphere near cover. Experiments with various anomaly test sets have shown that GAN-based anomaly detectors create a reliable anti-anomaly shield using the abovementioned assumptions. The proposed anomaly detector is tailored to the given classifier, but its limitation is due to the need for the availability of the database on which the classifier was trained. [ABSTRACT FROM AUTHOR]

Shadow removal for document images is an essential task for digitized document applications. Recent shadow removal models have been trained on pairs of shadow images and shadow-free images. However, obtaining a large, diverse dataset for document shadow removal takes time and effort. Thus, only small real datasets are available. Graphic renderers have been used to synthesize shadows to create relatively large datasets. However, the limited number of unique documents and the limited lighting environments adversely affect the network performance. This paper presents a large-scale, diverse dataset called the Synthetic Document with Diverse Shadows (SynDocDS) dataset. The SynDocDS comprises rendered images with diverse shadows augmented by a physics-based illumination model, which can be utilized to obtain a more robust and high-performance deep shadow removal network. In this paper, we further propose a Dual Shadow Fusion Network (DSFN). Unlike natural images, document images often have constant background colors requiring a high understanding of global color features for training a deep shadow removal network. The DSFN has a high global color comprehension and understanding of shadow regions and merges shadow attentions and features efficiently. We conduct experiments on three publicly available datasets, the OSR, Kligler's, and Jung's datasets, to validate our proposed method's effectiveness. In comparison to training on existing synthetic datasets, our model training on the SynDocDS dataset achieves an enhancement in the PSNR and SSIM, increasing them from 23.00 dB to 25.70 dB and 0.959 to 0.971 on average. In addition, the experiments demonstrated that our DSFN clearly outperformed other networks across multiple metrics, including the PSNR, the SSIM, and its impact on OCR performance. [ABSTRACT FROM AUTHOR]

The increasing growth in knowledge about the functioning of the nervous system of mammals and humans, as well as the significant neuromorphic technology developments in recent decades, has led to the emergence of a large number of brain–computer interfaces and neuroprosthetics for regenerative medicine tasks. Neurotechnologies have traditionally been developed for therapeutic purposes to help or replace motor, sensory or cognitive abilities damaged by injury or disease. They also have significant potential for memory enhancement. However, there are still no fully developed neurotechnologies and neural interfaces capable of restoring or expanding cognitive functions, in particular memory, in mammals or humans. In this regard, the search for new technologies in the field of the restoration of cognitive functions is an urgent task of modern neurophysiology, neurotechnology and artificial intelligence. The hippocampus is an important brain structure connected to memory and information processing in the brain. The aim of this paper is to propose an approach based on deep neural networks for the prediction of hippocampal signals in the CA1 region based on received biological input in the CA3 region. We compare the results of prediction for two widely used deep architectures: reservoir computing (RC) and long short-term memory (LSTM) networks. The proposed study can be viewed as a first step in the complex task of the development of a neurohybrid chip, which allows one to restore memory functions in the damaged rodent hippocampus. [ABSTRACT FROM AUTHOR]

Lithium-ion batteries are widely used in electric vehicles, energy storage power stations, and many other applications. Accurate and reliable monitoring of battery health status and remaining capacity is the key to establish a lithium-ion cell management system. In this paper, based on a Bayesian optimization algorithm, a deep neural network is structured to evaluate the whole charging curve of the battery using partial charging curve data as input. A 0.74 Ah battery is used for experiments, and the effect of different input data lengths is also investigated to check the high flexibility of the approach. The consequences show that using only 20 points of partial charging data as input, the whole charging profile of a cell can be exactly predicted with a root-mean-square error (RMSE) of less than 19.16 mAh (2.59% of the nominal capacity of 0.74 Ah), and its mean absolute percentage error (MAPE) is less than 1.84%. In addition, critical information including battery state-of-charge (SOC) and state-of-health (SOH) can be extracted in this way to provide a basis for safe and long-lasting battery operation. [ABSTRACT FROM AUTHOR]

Mainstream transferable adversarial attacks tend to introduce noticeable artifacts into the generated adversarial examples, which will impair the invisibility of adversarial perturbation and make these attacks less practical in real-world scenarios. To deal with this problem, in this paper, we propose a novel black-box adversarial attack method that can significantly improve the invisibility of adversarial examples. We analyze the sensitivity of a deep neural network in the frequency domain and take into account the characteristics of the human visual system in order to quantify the contribution of each frequency component in adversarial perturbation. Then, we collect a set of candidate frequency components that are insensitive to the human visual system by applying K-means clustering and we propose a joint loss function during the generation of adversarial examples, limiting the frequency distribution of perturbations during attacks. The experimental results show that the proposed method significantly outperforms existing transferable black-box adversarial attack methods in terms of invisibility, which verifies the superiority, applicability and potential of this work. [ABSTRACT FROM AUTHOR]

Real-time detection of workers is crucial in construction safety management. Deep learning-based detecting methods are valuable, but always challenged by the possibility of target missing or identity errors under complex scenarios. To address these limitations, previous research depended on re-training for new models or datasets, which are prohibitively time-consuming and incur high computing demands. However, we demonstrate that the better detecting model might not rely on more re-training of weights; instead, a training-free model can achieve even better performance by integrating head information. In this paper, a new head-detecting branch (55 MB) is added to the Keypoint Region-based Convolutional Network (Keypoint R-CNN, 226 MB) without altering its original weights, allowing for a less occluded head to aid in body detection. We also deployed motion information and anthropometric data through a post-processing module to calculate movement relationships. This study achieved an identity F1-score (IDF1) of 97.609%, recall (Rcll) of 98.173%, precision (Prcn) of 97.052%, and accuracy of 95.329% as a state-of-the-art (SOTA) method for worker detection. This exploration breaks the inertial attitudes of re-training dependency and accelerates the application of universal models, in addition to reducing the computational difficulty for most construction sites, especially in scenarios with an insufficient graphics processing unit (GPU). More importantly, this study can address occlusion challenges effectively in the worker detection field, making it of practical significance. [ABSTRACT FROM AUTHOR]

To achieve successful prevention of fire incidents originating from human activities, it is imperative to possess a thorough understanding. This paper introduces a machine learning approach, specifically utilizing deep neural networks (DNN), to develop predictive models for fire occurrence in Keelung City, Taiwan. It investigates ten factors across demographic, architectural, and economic domains through spatial analysis and thematic maps generated from geographic information system data. These factors are then integrated as inputs for the DNN model. Through 50 iterations, performance indices including the coefficient of determination (R2), root mean square error (RMSE), variance accounted for (VAF), prediction interval (PI), mean absolute error (MAE), weighted index (WI), weighted mean absolute percentage error (WMAPE), Nash–Sutcliffe efficiency (NS), and the ratio of performance to deviation (RPD) are computed, with average values of 0.89, 7.30 × 10−2, 89.21, 1.63, 4.90 × 10−2, 0.97, 2.92 × 10−1, 0.88, and 4.84, respectively. The model's predictions, compared with historical data, demonstrate its efficacy. Additionally, this study explores the impact of various urban renewal strategies using the DNN model, highlighting the significant influence of economic factors on fire incidence. This underscores the importance of economic factors in mitigating fire incidents and emphasizes their consideration in urban renewal planning. [ABSTRACT FROM AUTHOR]

Semi-supervised learning (SSL) models, integrating labeled and unlabeled data, have gained prominence in vision-based tasks, yet their susceptibility to adversarial attacks remains underexplored. This paper unveils the vulnerability of SSL models to gray-box adversarial attacks—a scenario where the attacker has partial knowledge of the model. We introduce an efficient attack method, Gray-box Adversarial Attack on Semi-supervised learning (GAAS), which exploits the dependency of SSL models on publicly available labeled data. Our analysis demonstrates that even with limited knowledge, GAAS can significantly undermine the integrity of SSL models across various tasks, including image classification, object detection, and semantic segmentation, with minimal access to labeled data. Through extensive experiments, we exhibit the effectiveness of GAAS, comparing it to white-box attack scenarios and underscoring the critical need for robust defense mechanisms. Our findings highlight the potential risks of relying on public datasets for SSL model training and advocate for the integration of adversarial training and other defense strategies to safeguard against such vulnerabilities. [ABSTRACT FROM AUTHOR]

LiDAR has high accuracy and resolution and is widely used in various fields. In particular, phase-modulated continuous-wave (PhMCW) LiDAR has merits such as low power, high precision, and no need for laser frequency modulation. However, with decreasing signal-to-noise ratio (SNR), the noise on the signal waveform becomes so severe that the current methods to extract the time-of-flight are no longer feasible. In this paper, a novel method that uses deep neural networks to measure the pulse width is proposed. The effects of distance resolution and SNR on the performance are explored. Recognition accuracy reaches 81.4% at a 0.1 m distance resolution and the SNR is as low as 2. We simulate a scene that contains a vehicle, a tree, a house, and a background located up to 6 m away. The reconstructed point cloud has good fidelity, the object contours are clear, and the features are restored. More precisely, the three distances are 4.73 cm, 6.00 cm, and 7.19 cm, respectively, showing that the performance of the proposed method is excellent. To the best of our knowledge, this is the first work that employs a neural network to directly process LiDAR signals and to extract their time-of-flight. [ABSTRACT FROM AUTHOR]

Deep neural networks excel in supervised learning tasks but are constrained by the need for extensive labeled data. Self-supervised learning emerges as a promising alternative, allowing models to learn without explicit labels. Information theory has shaped deep neural networks, particularly the information bottleneck principle. This principle optimizes the trade-off between compression and preserving relevant information, providing a foundation for efficient network design in supervised contexts. However, its precise role and adaptation in self-supervised learning remain unclear. In this work, we scrutinize various self-supervised learning approaches from an information-theoretic perspective, introducing a unified framework that encapsulates the self-supervised information-theoretic learning problem. This framework includes multiple encoders and decoders, suggesting that all existing work on self-supervised learning can be seen as specific instances. We aim to unify these approaches to understand their underlying principles better and address the main challenge: many works present different frameworks with differing theories that may seem contradictory. By weaving existing research into a cohesive narrative, we delve into contemporary self-supervised methodologies, spotlight potential research areas, and highlight inherent challenges. Moreover, we discuss how to estimate information-theoretic quantities and their associated empirical problems. Overall, this paper provides a comprehensive review of the intersection of information theory, self-supervised learning, and deep neural networks, aiming for a better understanding through our proposed unified approach. [ABSTRACT FROM AUTHOR]

DNN-based systems have demonstrated unprecedented performance in terms of accuracy and speed over the past decade. However, recent work has shown that such models may not be sufficiently robust during the inference process. Furthermore, due to the data-driven learning nature of DNNs, designing interpretable and generalizable networks is a major challenge, especially when considering critical applications such as medical computer-aided diagnostics (CAD) and other medical imaging tasks. Within this context, a line of approaches incorporating prior knowledge domain information into deep learning methods has recently emerged. In particular, many of these approaches utilize known physics-based forward imaging models, aimed at improving the stability and generalization ability of DNNs for medical imaging applications. In this paper, we review recent work focused on such physics-based or physics-prior-based learning for a variety of imaging modalities and medical applications. We discuss how the inclusion of such physics priors to the training process and/or network architecture supports their stability and generalization ability. Moreover, we propose a new physics-based approach, in which an explicit physics prior, which describes the relation between the input and output of the forward imaging model, is included as an additional input into the network architecture. Furthermore, we propose a tailored training process for this extended architecture, for which training data are generated with perturbed physical priors that are also integrated into the network. Within the scope of this approach, we offer a problem formulation for a regression task with a highly nonlinear forward model and highlight possible useful applications for this task. Finally, we briefly discuss future challenges for physics-informed deep learning in the context of medical imaging. [ABSTRACT FROM AUTHOR]

With the extensive adoption of generative models across various domains, the protection of copyright for these models has become increasingly vital. Some researchers suggest embedding watermarks in the images generated by these models as a means of preserving IP rights. In this paper, we find that existing generative model watermarking introduces high-frequency artifacts in the high-frequency spectrum of the marked images, thereby compromising the imperceptibility and security of the generative model watermarking system. Given this revelation, we propose an innovative image watermarking technology that takes into account frequency-domain imperceptibility. Our approach abandons the conventional convolutional neural network (CNN) structure typically used as the watermarking embedding network in popular watermarking techniques. This helps the image watermarking system to avoid the inherent high-frequency artifacts commonly produced by CNNs. In addition, we design a frequency perturbation generation network to generate low-frequency perturbations. These perturbations are subsequently added as watermarks to the low-frequency components of the carrier image, thus minimizing the impact of the watermark embedding process on the high-frequency properties of the image. The results show that our proposed watermarking framework can effectively embed low-frequency perturbation watermarks into images and effectively suppress high-frequency artifacts in images, thus significantly improving the frequency-domain imperceptibility and security of the image watermarking system. The introduced approach enhances the average invisibility performance in the frequency domain by up to 24.9 % when contrasted with prior methods. Moreover, the method attains superior image quality (>50 dB) in the spatial domain, accompanied by a 100 % success rate in watermark extraction in the absence of attacks. This underscores its capability to uphold the efficacy of the protected network and preserve the integrity of the watermarking process. It always maintains excellent imperceptibility and robustness. Thus, the framework shows great potential as a state-of-the-art solution for protecting intellectual property. [ABSTRACT FROM AUTHOR]

Artificial intelligence techniques are now widely used in various agricultural applications, including the detection of devastating diseases such as late blight (Phytophthora infestans) and early blight (Alternaria solani) affecting potato (Solanum tuberorsum L.) crops. In this paper, we present a mobile application for detecting potato crop diseases based on deep neural networks. The images were taken from the PlantVillage dataset with a batch of 1000 images for each of the three identified classes (healthy, early blight-diseased, late blight-diseased). An exploratory analysis of the architectures used for early and late blight diagnosis in potatoes was performed, achieving an accuracy of 98.7%, with MobileNetv2. Based on the results obtained, an offline mobile application was developed, supported on devices with Android 4.1 or later, also featuring an information section on the 27 diseases affecting potato crops and a gallery of symptoms. For future work, segmentation techniques will be used to highlight the damaged region in the potato leaf by evaluating its extent and possibly identifying different types of diseases affecting the same plant. [ABSTRACT FROM AUTHOR]

Three-dimensional (3D) models provide the most intuitive representation of geological conditions. Traditional modeling methods heavily depend on technicians' expertise and lack ease of updating. In this study, we introduce a deep learning-based method for 3D geological implicit modeling, leveraging a substantial dataset of geological drilling data. By applying resampling and normalization techniques, we standardize drilling data and significantly expand the dataset, making it suitable for training deep neural networks. Utilizing the characteristics of the sample data, we design and establish the network structure, loss function, and parameter configurations, resulting in the training of a deep neural network with high accuracy and robust generalization capability. Ultimately, we utilize the dataset generated from the network's predictions to render and construct the 3D geological model. The research in this paper demonstrates the significant promise of deep neural networks in addressing geological challenges. The deep learning-based implicit 3D modeling method surpasses traditional approaches in terms of generalization, convenience, and adaptability. [ABSTRACT FROM AUTHOR]

Cognitive decline represents a significant public health concern due to its severe implications on memory and general health. Early detection is crucial to initiate timely interventions and improve patient outcomes. However, traditional diagnosis methods often rely on personal interpretations or biases, may not detect the early stages of cognitive decline, or involve invasive screening procedures; thus, there is a growing interest in developing non-invasive methods benefiting also from the technological advances. Wearable devices and Internet of Things sensors can monitor various aspects of daily life together with health parameters and can provide valuable data regarding people's behavior. In this paper, we propose a technical solution that can be useful for potentially supporting cognitive decline assessment in early stages, by employing advanced machine learning techniques for detecting higher activity fragmentation based on daily activity monitoring using wearable devices. Our approach also considers data coming from wellbeing assessment questionnaires that can offer other important insights about a monitored person. We use deep neural network models to capture complex, non-linear relationships in the daily activities data and graph learning for the structural wellbeing information in the questionnaire answers. The proposed solution is evaluated in a simulated environment on a large synthetic dataset, the results showing that our approach can offer an alternative as a support for early detection of cognitive decline during patient-assessment processes. [ABSTRACT FROM AUTHOR]

Soybean (Glycine max (L.) Merr.) is a major source of oil and protein for human food and animal feed; however, soybean crops face diverse factors causing damage, including pathogen infections, environmental shifts, poor fertilization, and incorrect pesticide use, leading to reduced yields. Identifying the level of leaf damage aids yield projections, pesticide, and fertilizer decisions. Deep learning models (DLMs) and neural networks mastering tasks from abundant data have been used for binary healthy/unhealthy leaf classification. However, no DLM predicts and categorizes soybean leaf damage severity (five levels) for tailored pesticide use and yield forecasts. This paper introduces a novel DLM for accurate damage prediction and classification, trained on 2930 near-field soybean leaf images. The model quantifies damage severity, distinguishing healthy/unhealthy leaves and offering a comprehensive solution. Performance metrics include accuracy, precision, recall, and F1-score. This research presents a robust DLM for soybean damage assessment, supporting informed agricultural decisions based on specific damage levels and enhancing crop management and productivity. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have gained considerable attention for their expressive capabilities, but unfortunately they have serious robustness risks. Formal verification is an important technique to ensure network reliability. However, current verification techniques are unsatisfactory in time performance, which hinders the practical applications. To address this issue, we propose an efficient optimization method based on parallel acceleration with more computing resources. The method involves the speedup configuration of a partition-based verification aligned with the structures and robustness formal specifications of DNNs. A parallel verification framework is designed specifically for neural network verification systems, which integrates various auxiliary modules and accommodates diverse verification modes. The efficient parallel scheduling of verification queries within the framework enhances resource utilization and enables the system to process a substantial volume of verification tasks. We conduct extensive experiments on multiple commonly used verification benchmarks to demonstrate the rationality and effectiveness of the proposed method. The results show that higher efficiency is achieved after parallel optimization integration. [ABSTRACT FROM AUTHOR]

Battery state of health (SOH), which is a crucial parameter of the battery management system, reflects the rate of performance degradation and the aging level of lithium-ion batteries (LIBs) during operation. However, traditional machine learning models face challenges in accurately diagnosing battery SOH in complex application scenarios. Hence, we developed a deep learning framework for battery SOH estimation without prior knowledge of the degradation in battery capacity. Our framework incorporates a series of deep neural networks (DNNs) that utilize the direct current internal resistance (DCIR) feature to estimate the SOH. The correlation of the DCIR feature with the fade in capacity is quantified as strong under various conditions using Pearson correlation coefficients. We utilize the K-fold cross-validation method to select the hyperparameters in the DNN models and the optimal hyperparameter conditions compared with machine learning models with significant advantages and reliable prediction accuracies. The proposed algorithm is subjected to robustness validation, and the experimental results demonstrate that the model achieves reliable precision, with a mean absolute error (MAE) less than 0.768% and a root mean square error (RMSE) less than 1.185%, even when LIBs are subjected to varying application scenarios. Our study highlights the superiority and reliability of combining DNNs with DCIR features for battery SOH estimation. [ABSTRACT FROM AUTHOR]

Deep learning (DL) networks have shown attractive performance in medical image processing tasks such as brain tumor classification. However, they are often criticized as mysterious "black boxes". The opaqueness of the model and the reasoning process make it difficult for health workers to decide whether to trust the prediction outcomes. In this study, we develop an interpretable multi-part attention network (IMPA-Net) for brain tumor classification to enhance the interpretability and trustworthiness of classification outcomes. The proposed model not only predicts the tumor grade but also provides a global explanation for the model interpretability and a local explanation as justification for the proffered prediction. Global explanation is represented as a group of feature patterns that the model learns to distinguish high-grade glioma (HGG) and low-grade glioma (LGG) classes. Local explanation interprets the reasoning process of an individual prediction by calculating the similarity between the prototypical parts of the image and a group of pre-learned task-related features. Experiments conducted on the BraTS2017 dataset demonstrate that IMPA-Net is a verifiable model for the classification task. A percentage of 86% of feature patterns were assessed by two radiologists to be valid for representing task-relevant medical features. The model shows a classification accuracy of 92.12%, of which 81.17% were evaluated as trustworthy based on local explanations. Our interpretable model is a trustworthy model that can be used for decision aids for glioma classification. Compared with black-box CNNs, it allows health workers and patients to understand the reasoning process and trust the prediction outcomes. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have demonstrated remarkable performance in many areas but require significant computation and storage resources. Quantization is an effective method to reduce CNN complexity and implementation. The main research objective is to develop a scalable quantization algorithm for CNN hardware design and model the performance metrics for the purpose of CNN implementation in resource-constrained devices (RCDs) and optimizing layers in deep neural networks (DNNs). The algorithm novelty is based on blending two quantization techniques to perform full model quantization with optimum accuracy, and without additional neurons. The algorithm is applied to a selected CNN model and implemented on an FPGA. Implementing CNN using broad data is not possible due to capacity issues. With the proposed quantization algorithm, we succeeded in implementing the model on the FPGA using 16-, 12-, and 8-bit quantization. Compared to the 16-bit design, the 8-bit design offers a 44% decrease in resource utilization, and achieves power and energy reductions of 41% and 42%, respectively. Models show that trading off one quantization bit yields savings of approximately 5.4K LUTs, 4% logic utilization, 46.9 mW power, and 147 μ J energy. The models were also used to estimate performance metrics for a sample DNN design. [ABSTRACT FROM AUTHOR]

Recent research has delved into the biological parallels between deep neural networks (DNNs) in vision and human perception through the study of visual illusions. However, the bulk of this research is currently constrained to the investigation of visual illusions within a single model focusing on a singular type of illusion. There exists a need for a more comprehensive explanation of visual illusions in DNNs, as well as an expansion in the variety of illusions studied. This study is pioneering in its application of representational dissimilarity matrices and feature activation visualization techniques for a detailed examination of how five classic visual illusions are processed by DNNs. Our findings uncover the potential of DNNs to mimic human visual illusions, particularly highlighting notable differences in how these networks process illusions pertaining to color, contrast, length, angle, and spatial positioning. Although there are instances of consistency between DNNs and human perception in certain illusions, the performance distribution and focal points of interest within the models diverge from those of human observers. This study significantly advances our comprehension of DNNs' capabilities in handling complex visual tasks and their potential to emulate the human biological visual system. It also underscores the existing gaps in our understanding and processing of intricate visual information. While DNNs have shown progress in simulating human vision, their grasp of the nuance and intricacy of complex visual data still requires substantial improvement. [ABSTRACT FROM AUTHOR]

This research presents a comprehensive comparative analysis of SLAM algorithms and Deep Neural Network (DNN)-based Behavior Cloning (BC) navigation in outdoor agricultural environments. The study categorizes SLAM algorithms into laser-based and vision-based approaches, addressing the specific challenges posed by uneven terrain and the similarity between aisles in an orchard farm. The DNN-based BC navigation technique proves efficient, exhibiting reduced human intervention and providing a viable alternative for agricultural navigation. Despite the DNN-based BC navigation approach taking more time to reach its target due to a constant throttle limit for steady speed, the overall performance in terms of driving deviation and human intervention is notable compared to conventional SLAM algorithms. We provide comprehensive evaluation criteria for selecting optimal techniques for outdoor agricultural navigations. The algorithms were tested in three different scenarios: Precision, Speed, and Autonomy. Our proposed performance metric, P, is weighted and normalized. The DNN-based BC algorithm showed the best performance among the others, with a performance of 0.92 in the Precision and Autonomy scenarios. When Speed is more important, the RTAB-Map showed the best score with 0.96. In a case where Autonomy has a higher priority, Gmapping also showed a comparable performance of 0.92 with the DNN-based BC. [ABSTRACT FROM AUTHOR]

The structure and function of retinal vessels play a crucial role in diagnosing and treating various ocular and systemic diseases. Therefore, the accurate segmentation of retinal vessels is of paramount importance to assist a clinical diagnosis. U-Net has been highly praised for its outstanding performance in the field of medical image segmentation. However, with the increase in network depth, multiple pooling operations may lead to the problem of crucial information loss. Additionally, handling the insufficient processing of local context features caused by skip connections can affect the accurate segmentation of retinal vessels. To address these problems, we proposed a novel model for retinal vessel segmentation. The proposed model is implemented based on the U-Net architecture, with the addition of two blocks, namely, an MsFE block and MsAF block, between the encoder and decoder at each layer of the U-Net backbone. The MsFE block extracts low-level features from different scales, while the MsAF block performs feature fusion across various scales. Finally, the output of the MsAF block replaces the skip connection in the U-Net backbone. Experimental evaluations on the DRIVE dataset, CHASE_DB1 dataset, and STARE dataset demonstrated that MsAF-UNet exhibited excellent segmentation performance compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The past few years have witnessed a wider adoption of Internet of Things (IoT) devices. Since IoT devices are usually deployed in an open and uncertain environment, device authentication is of great importance. However, traditional device fingerprint (DF) extraction methods have several disadvantages. First, existing DF extraction methods need private information from devices to compute DFs, which puts the privacy of devices at stake. Second, the manually designing features-based methods suffer from poor performance. To tackle these limitations, we propose a Linear Residual Neural Network-based DF extraction method, Res-DFNN, which utilizes network packet data in the pcap file to generate DF. The key block is designed according to symmetry, and it is verified by simulation that our method achieves better performance in both non-private and privacy-preserving scenarios. [ABSTRACT FROM AUTHOR]

Variation in lighting conditions is a major cause of performance degradation in pattern recognition when using optical imaging. In this study, infrared (IR) and depth images were considered as possible robust alternatives against variations in illumination, particularly for improving the performance of automatic lip-reading. The variations due to lighting conditions were quantitatively analyzed for optical, IR, and depth images. Then, deep neural network (DNN)-based lip-reading rules were built for each image modality. Speech recognition techniques based on IR or depth imaging required an additional light source that emitted light in the IR range, along with a special camera. To mitigate this problem, we propose a method that does not use an IR/depth image directly, but instead estimates images based on the optical RGB image. To this end, a modified U-net was adopted to estimate the IR/depth image from an optical RGB image. The results show that the IR and depth images were rarely affected by the lighting conditions. The recognition rates for the optical, IR, and depth images were 48.29%, 95.76%, and 92.34%, respectively, under various lighting conditions. Using the estimated IR and depth images, the recognition rates were 89.35% and 80.42%, respectively. This was significantly higher than for the optical RGB images. [ABSTRACT FROM AUTHOR]

We explore to what extent data-driven prediction models have skills in forecasting daily sea-surface temperature (SST), which are comparable to or perform better than current physics-based operational systems over long-range forecast horizons. Three hybrid deep learning-based models are developed within the South China Sea (SCS) basin by integrating deep neural networks (back propagation, long short-term memory, and gated recurrent unit) with traditional empirical orthogonal function analysis and empirical mode decomposition. Utilizing a 40-year (1982–2021) satellite-based daily SST time series on a 0.25° grid, we train these models on the first 32 years (1982–2013) of detrended SST anomaly (SSTA) data. Their predictive accuracies are then validated using data from 2014 and tested over the subsequent seven years (2015–2021). The models' forecast skills are assessed using spatial anomaly correlation coefficient (ACC) and root-mean-square error (RMSE), with ACC proving to be a stricter metric. A forecast skill horizon, defined as the lead time before ACC drops below 0.6, is determined to be 50 days. The models are equally capable of achieving a basin-wide average ACC of ~0.62 and an RMSE of ~0.48 °C at this horizon, indicating a 36% improvement in RMSE over climatology. This implies that on average the forecast skill horizon for these models is beyond the available forecast length. Analysis of one model, the BP neural network, reveals a variable forecast skill horizon (5 to 50 days) for each individual day, showing that it can adapt to different time scales. This adaptability seems to be influenced by a number of mechanisms arising from the evident regional and global atmosphere–ocean coupling variations on time scales ranging from intraseasonal to decadal in the SSTA of the SCS basin. [ABSTRACT FROM AUTHOR]

To ensure precise and real-time perception of high-speed roadway conditions and minimize the potential threats to traffic safety posed by road debris and defects, this study designed a real-time monitoring and early warning system for high-speed road surface anomalies. Initially, an autonomous mobile intelligent road inspection robot, mountable on highway guardrails, along with a corresponding cloud-based warning platform, was developed. Subsequently, an enhanced target detection algorithm, YOLOv5s-L-OTA, was proposed. Incorporating GSConv for lightweight improvements to standard convolutions and employing the optimal transport assignment for object detection (OTA) strategy, the algorithm's robustness in multi-object label assignment was enhanced, significantly improving both model accuracy and processing speed. Ultimately, this refined algorithm was deployed on the intelligent inspection robot and validated in real-road environments. The experimental results demonstrated the algorithm's effectiveness, significantly boosting the capability for real-time, precise detection of high-speed road surface anomalies, thereby ensuring highway safety and substantially reducing the risk of liability disputes and personal injuries. [ABSTRACT FROM AUTHOR]

Monitoring soil organic carbon (SOC) typically assumes conducting a labor-intensive soil sampling campaign, followed by laboratory testing, which is both expensive and impractical for generating useful, spatially continuous data products. The present study leverages the power of machine learning (ML) and, in particular, deep neural networks (DNNs) for segmentation, as well as satellite imagery, to estimate the SOC remotely. We propose a new two-stage pipeline for remote SOC estimation, which relies on using a DNN trained to classify land cover to perform feature extraction, while the SOC estimation is performed by a different ML model. The first stage is an image segmentation DNN with the U-Net architecture, which is trained to estimate the land cover for an observed geographical region, based on multi-spectral images taken by the Sentinel-2 satellite constellation. This estimator is subsequently used to extract the latent feature vector for each of the output pixels, by rolling back from the output (dense) layer of the U-Net and accessing the last available convolutional layer of the same dimension as our desired output. The second stage is trained on a set of feature vectors extracted at the coordinates for which manual SOC measurements exist. We tested a variety of ML models and report on their performance. Using the best extremely randomized trees model, we generated a spatially continuous map of SOC estimations for the region of Tuscany, in Italy, with a resolution of 10 m, to share with the researchers as a means of validating the results and to demonstrate the efficiency of the proposed approach, which can can easily be scaled to create a global continuous SOC map. [ABSTRACT FROM AUTHOR]

Electric vehicles are widely adopted globally as a sustainable mode of transportation. With the increased availability of onboard computation and communication capabilities, vehicles are moving towards automated driving and intelligent transportation systems. The adaption of technologies such as IoT, edge intelligence, 5G, and blockchain in vehicle architecture has increased possibilities towards efficient and sustainable transportation systems. In this article, we present a comprehensive study and analysis of the edge computing paradigm, explaining elements of edge AI. Furthermore, we discussed the edge intelligence approach for deploying AI algorithms and models on edge devices, which are typically resource-constrained devices located at the edge of the network. It mentions the advantages of edge intelligence and its use cases in smart electric vehicles. It also discusses challenges and opportunities and provides in-depth analysis for optimizing computation for edge intelligence. Finally, it sheds some light on the research roadmap on AI for edge and AI on edge by dividing efforts into topology, content, service segments, model adaptation, framework design, and processor acceleration, all of which stand to gain advantages from AI technologies. Investigating the incorporation of important technologies, issues, opportunities, and Roadmap in this study will be a valuable resource for the community engaged in research on edge intelligence in electric vehicles. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy is a prevalent eye disease that poses a potential risk of blindness. Nevertheless, due to the small size of diabetic retinopathy lesions and the high interclass similarity in terms of location, color, and shape among different lesions, the segmentation task is highly challenging. To address these issues, we proposed a novel framework named nmODE-Unet, which is based on the nmODE (neural memory Ordinary Differential Equation) block and U-net backbone. In nmODE-Unet, the shallow features serve as input to the nmODE block, and the output of the nmODE block is fused with the corresponding deep features. Extensive experiments were conducted on the IDRiD dataset, e_ophtha dataset, and the LGG segmentation dataset, and the results demonstrate that, in comparison to other competing models, nmODE-Unet showcases a superior performance. [ABSTRACT FROM AUTHOR]

In the era of artificial intelligence (AI), the deployment of intelligent systems for autonomous decision making has surged across diverse fields. However, the widespread adoption of AI technology is hindered by the risks associated with ceding control to autonomous systems, particularly in critical domains. Explainable artificial intelligence (XAI) has emerged as a critical subdomain fostering human understanding and trust. It addresses the opacity of complex models such as vision transformers (ViTs), which have gained prominence lately. With the expanding landscape of XAI methods, selecting the most effective method remains an open question, due to the lack of a ground-truth label for explainability. To avoid subjective human judgment, numerous metrics have been developed, with each aiming to fulfill certain properties required for a valid explanation. This study conducts a detailed evaluation of various XAI methods applied to the ViT architecture, thereby exploring metrics criteria like faithfulness, coherence, robustness, and complexity. We especially study the metric convergence, correlation, discriminative power, and inference time of both XAI methods and metrics. Contrary to expectations, the metrics of each criterion reveal minimal convergence and correlation. This study not only challenges the conventional practice of metric-based ranking of XAI methods but also underscores the dependence of explanations on the experimental environment, thereby presenting crucial considerations for the future development and adoption of XAI methods in real-world applications. [ABSTRACT FROM AUTHOR]