Search

This paper presents a computational approach for quantifying soybean defects through seed classification using deep learning techniques. To differentiate between good and defective soybean seeds quickly and accurately, we introduce a lightweight soybean seed defect identification network (SSDINet). Initially, the labeled soybean seed dataset is developed and processed through the proposed seed contour detection (SCD) algorithm, which enhances the quality of soybean seed images and performs segmentation, followed by SSDINet. The classification network, SSDINet, consists of a convolutional neural network, depthwise convolution blocks, and squeeze-and-excitation blocks, making the network lightweight, faster, and more accurate than other state-of-the-art approaches. Experimental results demonstrate that SSDINet achieved the highest accuracy, of 98.64%, with 1.15 M parameters in 4.70 ms, surpassing existing state-of-the-art models. This research contributes to advancing deep learning techniques in agricultural applications and offers insights into the practical implementation of seed classification systems for quality control in the soybean industry. [ABSTRACT FROM AUTHOR]

Recent advances in intelligent wearable devices have brought tremendous chances for the development of healthcare monitoring system. However, the data collected by various sensors in it are user-privacy-related information. Once the individuals' privacy is subjected to attacks, it can potentially cause serious hazards. For this reason, a feasible solution built upon the compression-encryption architecture is proposed. In this scheme, we design an Adaptive Sparse Basis Compressive Sensing (ASB-CS) model by leveraging Singular Value Decomposition (SVD) manipulation, while performing a rigorous proof of its effectiveness. Additionally, incorporating the Parametric Deformed Exponential Rectified Linear Unit (PDE-ReLU) memristor, a new fractional-order Hopfield neural network model is introduced as a pseudo-random number generator for the proposed cryptosystem, which has demonstrated superior properties in many aspects, such as hyperchaotic dynamics and multistability. To be specific, a plain medical image is subjected to the ASB-CS model and bidirectional diffusion manipulation under the guidance of the key-controlled cipher flows to yield the corresponding cipher image without visual semantic features. Ultimately, the simulation results and analysis demonstrate that the proposed scheme is capable of withstanding multiple security attacks and possesses balanced performance in terms of compressibility and robustness. [ABSTRACT FROM AUTHOR]

Wireless Sensor Networks (WSNs) and the Internet of Things (IoT) have emerged as transforming technologies, bringing the potential to revolutionize a wide range of industries such as environmental monitoring, agriculture, manufacturing, smart health, home automation, wildlife monitoring, and surveillance. Population expansion, changes in the climate, and resource constraints all offer problems to modern IoT applications. To solve these issues, the integration of Wireless Sensor Networks (WSNs) and the Internet of Things (IoT) has come forth as a game-changing solution. For example, in agricultural environment, IoT-based WSN has been utilized to monitor yield conditions and automate agriculture precision through different sensors. These sensors are used in agriculture environments to boost productivity through intelligent agricultural decisions and to collect data on crop health, soil moisture, temperature monitoring, and irrigation. However, sensors have finite and non-rechargeable batteries, and memory capabilities, which might have a negative impact on network performance. When a network is distributed over a vast area, the performance of WSN-assisted IoT suffers. As a result, building a stable and energy-efficient routing infrastructure is quite challenging in order to extend network lifetime. To address energy-related issues in scalable WSN-IoT environments for future IoT applications, this research proposes EEDC: An Energy Efficient Data Communication scheme by utilizing "Region based Hierarchical Clustering for Efficient Routing (RHCER)"—a multi-tier clustering framework for energy-aware routing decisions. The sensors deployed for IoT application data collection acquire important data and select cluster heads based on a multi-criteria decision function. Further, to ensure efficient long-distance communication along with even load distribution across all network nodes, a subdivision technique was employed in each tier of the proposed framework. The proposed routing protocol aims to provide network load balancing and convert communicating over long distances into shortened multi-hop distance communications, hence enhancing network lifetime.The performance of EEDC is compared to that of some existing energy-efficient protocols for various parameters. The simulation results show that the suggested methodology reduces energy usage by almost 31% in sensor nodes and provides almost 38% improved packet drop ratio. [ABSTRACT FROM AUTHOR]

Intrusion detection systems, also known as IDSs, are widely regarded as one of the most essential components of an organization's network security. This is because IDSs serve as the organization's first line of defense against several cyberattacks and are accountable for accurately detecting any possible network intrusions. Several implementations of IDSs accomplish the detection of potential threats throughout flow-based network traffic analysis. Traditional IDSs frequently struggle to provide accurate real-time intrusion detection while keeping up with the changing landscape of threat. Innovative methods used to improve IDSs' performance in network traffic analysis are urgently needed to overcome these drawbacks. In this study, we introduced a model called a deep neural decision forest (DNDF), which allows the enhancement of classification trees with the power of deep networks to learn data representations. We essentially utilized the CICIDS 2017 dataset for network traffic analysis and extended our experiments to evaluate the DNDF model's performance on two additional datasets: CICIDS 2018 and a custom network traffic dataset. Our findings showed that DNDF, a combination of deep neural networks and decision forests, outperformed reference approaches with a remarkable precision of 99.96% by using the CICIDS 2017 dataset while creating latent representations in deep layers. This success can be attributed to improved feature representation, model optimization, and resilience to noisy and unbalanced input data, emphasizing DNDF's capabilities in intrusion detection and network security solutions. [ABSTRACT FROM AUTHOR]

The comparison of low-rank-based learning models for multi-label categorization of attacks for intrusion detection datasets is presented in this work. In particular, we investigate the performance of three low-rank-based machine learning (LR-SVM) and deep learning models (LR-CNN), (LR-CNN-MLP) for classifying intrusion detection data: Low Rank Representation (LRR) and Non-negative Low Rank Representation (NLR). We also look into how these models' performance is affected by hyperparameter tweaking by using Guassian Bayes Optimization. The tests has been run on merging two intrusion detection datasets that are available to the public such as BoT-IoT and UNSW- NB15 and assess the models' performance in terms of key evaluation criteria, including precision, recall, F1 score, and accuracy. Nevertheless, all three models perform noticeably better after hyperparameter modification. The selection of low-rank-based learning models and the significance of the hyperparameter tuning log for multi-label classification of intrusion detection data have been discussed in this work. A hybrid security dataset is used with low rank factorization in addition to SVM, CNN and CNN-MLP. The desired multilabel results have been obtained by considering binary and multi-class attack classification as well. Low rank CNN-MLP achieved suitable results in multilabel classification of attacks. Also, a Gaussian-based Bayesian optimization algorithm is used with CNN-MLP for hyperparametric tuning and the desired results have been achieved using c and γ for SVM and α and β for CNN and CNN-MLP on a hybrid dataset. The results show the label UDP is shared among analysis, DoS and shellcode. The accuracy of classifying UDP among three classes is 98.54%. [ABSTRACT FROM AUTHOR]

Remote Sensing (RS) images are usually captured at resolutions lower than those required. Deep Learning (DL)-based super-resolution (SR) architectures are typically used to increase the resolution artificially. In this study, we designed a new architecture called TESR (Two-stage approach for Enhancement and super-resolution), leveraging the power of Vision Transformers (ViT) and the Diffusion Model (DM) to increase the resolution of RS images artificially. The first stage is the ViT-based model, which serves to increase resolution. The second stage is an iterative DM pre-trained on a larger dataset, which serves to increase image quality. Every stage is trained separately on the given task using a separate dataset. The self-attention mechanism of the ViT helps the first stage generate global and contextual details. The iterative Diffusion Model helps the second stage enhance the image's quality and generate consistent and harmonic fine details. We found that TESR outperforms state-of-the-art architectures on super-resolution of remote sensing images on the UCMerced benchmark dataset. Considering the PSNR/SSIM metrics, TESR improves SR image quality as compared to state-of-the-art techniques from 34.03/0.9301 to 35.367/0.9449 in the scale ×2. On a scale of ×3, it improves from 29.92/0.8408 to 32.311/0.91143. On a scale of ×4, it improves from 27.77/0.7630 to 31.951/0.90456. We also found that the Charbonnier loss outperformed other loss functions in the training of both stages of TESR. The improvement was by a margin of 21.5%/14.3%, in the PSNR/SSIM, respectively. The source code of TESR is open to the community. [ABSTRACT FROM AUTHOR]

In many cases, images contain sensitive information and patterns that require secure processing to avoid risk. It can be accessed by unauthorized users who can illegally exploit them to threaten the safety of people's life and property. Protecting the privacies of the images has quickly become one of the biggest obstacles that prevent further exploration of image data. In this paper, we propose a novel privacy-preserving scheme to protect sensitive information within images. The proposed approach combines deoxyribonucleic acid (DNA) sequencing code, Arnold transformation (AT), and a chaotic dynamical system to construct an initial S-box. Various tests have been conducted to validate the randomness of this newly constructed S-box. These tests include National Institute of Standards and Technology (NIST) analysis, histogram analysis (HA), nonlinearity analysis (NL), strict avalanche criterion (SAC), bit independence criterion (BIC), bit independence criterion strict avalanche criterion (BIC-SAC), bit independence criterion nonlinearity (BIC-NL), equiprobable input/output XOR distribution, and linear approximation probability (LP). The proposed scheme possesses higher security wit NL = 103.75, SAC ≈ 0.5 and LP = 0.1560. Other tests such as BIC-SAC and BIC-NL calculated values are 0.4960 and 112.35, respectively. The results show that the proposed scheme has a strong ability to resist many attacks. Furthermore, the achieved results are compared to existing state-of-the-art methods. The comparison results further demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The rapid increase in population and transportation resources presents numerous challenges, including traffic congestion and accidents. This paper proposes a smart traffic management (STM) framework that combines the Internet of Vehicles (IoV) and game theory to manage traffic loads at road intersections. The intersection is considered a non-cooperative game, where traffic flow for each route is determined by the Nash Equilibrium (NE) to ensure that no individual can improve their performance by changing their strategy. In severe congestion, many players/vehicles significantly affect the strategy selection process. To address this, we adopt an agent-representative approach, using spectral clustering to group players with the same strategies and payoff. The proposed STM is compared with other schemes, i.e., the conventional approach without NE and SmartRoute. Our STM significantly outperformed existing protocols, reducing intersection traffic intensity by 30% and average waiting time by 40%, demonstrating its effectiveness in managing traffic loads at uncontrolled intersections. [Display omitted] • In STM, intelligent vehicles compete for optimal rewards by selecting suitable routes for shorter travel time. • STM uses Nash Equilibrium (NE) to balance traffic load and improve flow on all strategies. • STM is divided into (1) real-time data collection and intersection modeling, and (2) representative agent/CH selection to reduce opponents. Spectral clustering is used for cluster formation and agent/CH selection. • The proposed scheme is evaluated against SmartRoute, MCDP, and VANET-based A* for performance comparison. [ABSTRACT FROM AUTHOR]

Smart traffic engineering and intelligent transportation services are in increasing demand from governmental authorities to optimize traffic performance and thus reduce energy costs, increase the drivers' safety and comfort, ensure traffic laws enforcement, and detect traffic violations. In this paper, we address this challenge, and we leverage the use of Artificial Intelligence (AI) and Unmanned Aerial Vehicles (UAVs) to develop an AI-integrated video analytics framework, called TAU (Traffic Analysis from UAVs), for automated traffic analytics and understanding. Unlike previous works on traffic video analytics, we propose an automated object detection and tracking pipeline from video processing to advanced traffic understanding using high-resolution UAV images. TAU combines six main contributions. First, it proposes a pre-processing algorithm to adapt the high-resolution UAV image as input to the object detector without lowering the resolution. This ensures an excellent detection accuracy from high-quality features, particularly the small size of detected objects from UAV images. Second, it introduces an algorithm for recalibrating the vehicle coordinates to ensure that vehicles are uniquely identified and tracked across the multiple crops of the same frame. Third, it presents a speed calculation algorithm based on accumulating information from successive frames. Fourth, TAU counts the number of vehicles per traffic zone based on the Ray Tracing algorithm. Fifth, TAU has a fully independent algorithm for crossroad arbitration based on the data gathered from the different zones surrounding it. Sixth, TAU introduces a set of algorithms for extracting twenty-four types of insights from the raw data collected. These insights facilitate the traffic understanding using curves, histograms, heatmaps, and animations. The present work presents a valuable added value for academic researchers and transportation engineers to automate the traffic video analytics process and extract useful insights to optimize traffic performance. TAU is a ready-to-use framework for any Transportation Engineer to better understand and manage daily road traffic (video demonstrations are provided here: https://youtu.be/wXJV0H7LviU and here: https://youtu.be/kGv0gmtVEbI). The source code is available at: https://github.com/bilel-bj/TAU. [ABSTRACT FROM AUTHOR]

Alzheimer's disease is a severe brain disorder that causes harm in various brain areas and leads to memory damage. The limited availability of labeled medical data poses a significant challenge for accurate Alzheimer's disease detection. There is a critical need for effective methods to improve the accuracy of Alzheimer's disease detection, considering the scarcity of labeled data, the complexity of the disease, and the constraints related to data privacy. To address this challenge, our study leverages the power of Big Data in the form of pre-trained Convolutional Neural Networks (CNNs) within the framework of Few-Shot Learning (FSL) and ensemble learning. We propose an ensemble approach based on a Prototypical Network (ProtoNet), a powerful method in FSL, integrating various pre-trained CNNs as encoders. This integration enhances the richness of features extracted from medical images. Our approach also includes a combination of class-aware loss and entropy loss to ensure a more precise classification of Alzheimer's disease progression levels. The effectiveness of our method was evaluated using two datasets, the Kaggle Alzheimer dataset, and the ADNI dataset, achieving an accuracy of 99.72% and 99.86%, respectively. The comparison of our results with relevant state-of-the-art studies demonstrated that our approach achieved superior accuracy and highlighted its validity and potential for real-world applications in early Alzheimer's disease detection.

Semantic segmentation of Remote Sensing (RS) images involves the classification of each pixel in a satellite image into distinct and non-overlapping regions or segments. This task is crucial in various domains, including land cover classification, autonomous driving, and scene understanding. While deep learning has shown promising results, there is limited research that specifically addresses the challenge of processing fine details in RS images while also considering the high computational demands. To tackle this issue, we propose a novel approach that combines convolutional and transformer architectures. Our design incorporates convolutional layers with a low receptive field to generate fine-grained feature maps for small objects in very high-resolution images. On the other hand, transformer blocks are utilized to capture contextual information from the input. By leveraging convolution and self-attention in this manner, we reduce the need for extensive downsampling and enable the network to work with full-resolution features, which is particularly beneficial for handling small objects. Additionally, our approach eliminates the requirement for vast datasets, which is often necessary for purely transformer-based networks. In our experimental results, we demonstrate the effectiveness of our method in generating local and contextual features using convolutional and transformer layers, respectively. Our approach achieves a mean dice score of 80.41%, outperforming other well-known techniques such as UNet, Fully-Connected Network (FCN), Pyramid Scene Parsing Network (PSP Net), and the recent Convolutional vision Transformer (CvT) model, which achieved mean dice scores of 78.57%, 74.57%, 73.45%, and 62.97% respectively, under the same training conditions and using the same training dataset., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2024 The Author(s).)

Since the release of ChatGPT, numerous studies have highlighted the remarkable performance of ChatGPT, which often rivals or even surpasses human capabilities in various tasks and domains. However, this paper presents a contrasting perspective by demonstrating an instance where human performance excels in typical tasks suited for ChatGPT, specifically in the domain of computer programming. We utilize the IEEExtreme Challenge competition as a benchmark-a prestigious, annual international programming contest encompassing a wide range of problems with different complexities. To conduct a thorough evaluation, we selected and executed a diverse set of 102 challenges, drawn from five distinct IEEExtreme editions, using three major programming languages: Python, Java, and C++. Our empirical analysis provides evidence that contrary to popular belief, human programmers maintain a competitive edge over ChatGPT in certain aspects of problem-solving within the programming context. In fact, we found that the average score obtained by ChatGPT on the set of IEEExtreme programming problems is 3.9 to 5.8 times lower than the average human score, depending on the programming language. This paper elaborates on these findings, offering critical insights into the limitations and potential areas of improvement for AI-based language models like ChatGPT., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2023 The Authors.)

Having access to safe water and using it properly is crucial for human well-being, sustainable development, and environmental conservation. Nonetheless, the increasing disparity between human demands and natural freshwater resources is causing water scarcity, negatively impacting agricultural and industrial efficiency, and giving rise to numerous social and economic issues. Understanding and managing the causes of water scarcity and water quality degradation are essential steps toward more sustainable water management and use. In this context, continuous Internet of Things (IoT)-based water measurements are becoming increasingly crucial in environmental monitoring. However, these measurements are plagued by uncertainty issues that, if not handled correctly, can introduce bias and inaccuracy into our analysis, decision-making processes, and results. To cope with uncertainty issues related to sensed water data, we propose combining network representation learning with uncertainty handling methods to ensure rigorous and efficient modeling management of water resources. The proposed approach involves accounting for uncertainties in the water information system by leveraging probabilistic techniques and network representation learning. It creates a probabilistic embedding of the network, enabling the classification of uncertain representations of water information entities, and applies evidence theory to enable decision making that is aware of uncertainties, ultimately choosing appropriate management strategies for affected water areas.

Extensive research has been conducted on healthcare technology and service advancements during the last decade. The Internet of Medical Things (IoMT) has demonstrated the ability to connect various medical apparatus, sensors, and healthcare specialists to ensure the best medical treatment in a distant location. Patient safety has improved, healthcare prices have decreased dramatically, healthcare services have become more approachable, and the operational efficiency of the healthcare industry has increased. This research paper offers a recent review of current and future healthcare applications, security, market trends, and IoMT-based technology implementation. This research paper analyses the advancement of IoMT implementation in addressing various healthcare concerns from the perspectives of enabling technologies, healthcare applications, and services. The potential obstacles and issues of the IoMT system are also discussed. Finally, the survey includes a comprehensive overview of different disciplines of IoMT to empower future researchers who are eager to work on and make advances in the field to obtain a better understanding of the domain., Competing Interests: Conflict of interestThe authors share no conflict of interests., (© The Author(s) 2022, corrected publication 2022.)