In recent years, the utilization of unmanned aerial vehicles (UAVs) for aerial target detection has gained significant attention due to their high-altitude perspective and maneuverability, which offer novel opportunit...
详细信息
To solve the problems of vote forgery and malicious election of candidate nodes in the Raft consensus algorithm, we combine zero trust with the Raft consensus algorithm and propose a secure and efficient consensus alg...
详细信息
People-centric activity recognition is one of the most critical technologies in a wide range of real-world applications,including intelligent transportation systems, healthcare services, and brain-computer interfaces....
详细信息
People-centric activity recognition is one of the most critical technologies in a wide range of real-world applications,including intelligent transportation systems, healthcare services, and brain-computer interfaces. Large-scale data collection and annotation make the application of machine learning algorithms prohibitively expensive when adapting to new tasks. One way of circumventing this limitation is to train the model in a semi-supervised learning manner that utilizes a percentage of unlabeled data to reduce the labeling burden in prediction tasks. Despite their appeal, these models often assume that labeled and unlabeled data come from similar distributions, which leads to the domain shift problem caused by the presence of distribution gaps. To address these limitations, we propose herein a novel method for people-centric activity recognition,called domain generalization with semi-supervised learning(DGSSL), that effectively enhances the representation learning and domain alignment capabilities of a model. We first design a new autoregressive discriminator for adversarial training between unlabeled and labeled source domains, extracting domain-specific features to reduce the distribution gaps. Second, we introduce two reconstruction tasks to capture the task-specific features to avoid losing information related to representation learning while maintaining task-specific consistency. Finally, benefiting from the collaborative optimization of these two tasks, the model can accurately predict both the domain and category labels of the source domains for the classification task. We conduct extensive experiments on three real-world sensing datasets. The experimental results show that DGSSL surpasses the three state-of-the-art methods with better performance and generalization.
In recent years, remote sensing object detection has become a research hotspot in computer vision tasks. However, previous approaches for remote sensing object detection often overlook the rich contextual information ...
详细信息
As deep learning advances, neural network technologies are increasingly penetrating the field of steel surface defect detection. To tackle the challenges of low accuracy and inadequate quality, we introduce CMS-YOLOv8...
详细信息
作者：
Gudadhe, Amit AnilReddy, K.T.V.
Faculty of Engineering and Technology Computer Science and Design Department Wardha India
Faculty of Engineering and Technology Wardha India
For social and economic development of any region, groundwater plays a very vital role. Surface water infiltration depends on various parameters of the earth. The parameters includes Slope, Geology, Soil Type, Land Us...
详细信息
Underwater target detection is an important method for detecting marine organisms. However, due to the image occlusion of underwater targets, blurred water quality, poor lighting conditions, small targets, and complex...
详细信息
Photovoltaic(PV)arrays are usually installed in open areas;hence,they are vulnerable to lightning strikes that can result in cell degradation,complete damage,service disruption,and increased maintenance *** a result,i...
详细信息
Photovoltaic(PV)arrays are usually installed in open areas;hence,they are vulnerable to lightning strikes that can result in cell degradation,complete damage,service disruption,and increased maintenance *** a result,it is imperative to develop an effective and efficient lightning protection system by evaluating the transient behaviour of PV arrays during lightning *** aim is to evaluate the transient analysis of large-scale PV systems when subjected to lightning strikes using the finite difference time domain(FDTD)*** overvoltages are calculated at various points within the mounting *** optimise the FDTD method's execution time and make it more suitable for less powerful hardware,a variable cell size approach is ***,larger cell dimensions are used in the earthing system and smaller cell dimensions are used in the mounting *** FDTD method is utilised to calculate the temporal variation of transient overvoltages for large-scale PV systems under different scenarios,including variations in the striking point,soil resistivity,and the presence of a metal *** results indicate that the highest transient overvoltages occur at the striking point,and these values increase with the presence of a PV metal frame as well as with higher soil ***,a comparison is performed between the overvoltage results obtained from the FDTD approach and the partial element equivalent circuit(PEEC)method at the four corner points of the mounting systems to demonstrate the superior accuracy of the FDTD ***,a laboratory experiment is conducted on a small-scale PV system to validate the simulation *** calculated overvoltages obtained from the FDTD and PEEC methods are compared with the measured values,yielding a mean absolute error of 5%and 11%for the FDTD and PEEC methods,respectively,thereby confirming the accuracy of the FDTD simulation model.
The advent of the Internet of Things (IoT) has revolutionized connectivity by interconnecting a vast array of devices, underscoring the critical need for robust data security, particularly at the Physical Layer Securi...
详细信息
The advent of the Internet of Things (IoT) has revolutionized connectivity by interconnecting a vast array of devices, underscoring the critical need for robust data security, particularly at the Physical Layer Security (PLS). Ensuring data confidentiality and integrity during wireless communications poses a primary challenge in IoT environments. Additionally, within the constrained frequency bands available, Cognitive Radio Networks (CRNs) has emerged as an urgent necessity to optimize spectrum utilization. This technology enables intelligent management of radio frequencies, enhancing network efficiency and adaptability to dynamic environmental changes. In this research, we focus on examining the PLS for the primary channel within the underlying CRNs. Our proposed model involves a primary source-destination pair and a secondary transmitter-receiver pair sharing the same frequency band simultaneously. In the presence of a common eavesdropper, the primary concern lies in securing the primary link communication. The secondary user (SU) acts as cooperative jamming, strategically allocating a portion of its transmission power to transmit artificial interference, thus confusing the eavesdropper and protecting the primary user's (PU) communication. The transmit power of the SU is regulated by the maximum interference power tolerated by the primary network's receiver. To evaluate the effectiveness of our proposed protocol, we develop closed-form mathematical expressions for intercept probability ((Formula presented.)) and outage probability (OP) along the primary communication link. Additionally, we derive mathematical expressions for OP along the secondary communications network. Furthermore, we investigate the impact of transmit power allocation on intercept and outage probabilities across various links. Through both simulation and theoretical analysis, our protocol aims to enhance protection and outage efficiency for the primary link while ensuring appropriate secondary
In today's society, people increasingly need information acquisition due to the rapid development of science and technology and the consequent increase in available data. However, finding the information users nee...
详细信息
暂无评论