This work presents a fault diagnoser for a three-wheel omnidirectional vehicle, which is based on the FDI (Fault Detection and Isolation) approach proposed by Massoumnia [1]. It is designed for a linearized model of t...
详细信息
Video-based anomaly detection opens up numerous practical applications such as video surveillance and healthcare. Despite having been studied for an extended period, this problem remains highly challenging because of ...
Video-based anomaly detection opens up numerous practical applications such as video surveillance and healthcare. Despite having been studied for an extended period, this problem remains highly challenging because of variations in anomalies caused by people, objects, or contexts. Furthermore, the issue of missing annotated datasets and imbalances in the training data can complicate the development of automatic detection systems. Moreover, many existing systems have attempted to detect anomalies by performing binary classification without providing any explanation of the results. This paper proposes a method to detect anomalies from video while providing a semantic explanation of the cause. We first deploy off-the-shelf techniques to detect people and objects in the scene, then we extract semantic features (motion, appearance, posture) to recognize anomalies. To focus more on posture change, we introduce new postural descriptions and time series descriptions of poses using a spatiotemporal graph convolutional neural network. The proposed method is validated on three public datasets: UCSD Ped2, CUHK Avenue, and ShanghaiTech Campus showing improvements over the baseline methods.
Successful adoption of distributed clean energy resources requires the enhancement of system flexibility from both the generation and the demand side. This paper presents a case study on the Punggol Digital District o...
详细信息
One of the challenges in ensuring the security of Internet of Things (IoT) devices involves preserving control flow integrity (CFI) against return-oriented programming (ROP) attacks. Traditional defense mechanisms, wh...
One of the challenges in ensuring the security of Internet of Things (IoT) devices involves preserving control flow integrity (CFI) against return-oriented programming (ROP) attacks. Traditional defense mechanisms, while effective, often require substantial changes to the compiler, thereby limiting their applicability to a broad range of devices. To overcome this challenge, we present the Parity Shadow Stack — a novel, compact shadow stack mechanism that leverages dynamic instrumentation and binary rewriting to achieve similar effects compared to compiler-based methods. Further, this method does not require application source code and can work with legacy applications where only the binary is available. Finally, we demonstrate that it is more flexible as it provides a programming interface to add additional security semantics checks. Through our evaluations, we find that our method introduces a remarkably low overhead, less than 10 percent for medium-size or large-size applications. Therefore, we conclude it provides robust and practical defense against ROP attacks without requiring source code or compiler changes, nor does it require any specialized hardware.
Human big data decoding is of great potential to reveal the complex patterns of human dynamics like physiological and biomechanical signals. In this study, we take special interest in brain visual dynamics, e.g., eye ...
详细信息
ISBN:
(纸本)9781728111797
Human big data decoding is of great potential to reveal the complex patterns of human dynamics like physiological and biomechanical signals. In this study, we take special interest in brain visual dynamics, e.g., eye movement signals, and investigate how to leverage eye signal decoding to provide a voice-free communication possibility for ALS patients who lose ability to control their muscles. Due to substantial complexity of visual dynamics, we propose a deep learning framework to decode the visual dynamics when the user performs eye-writing tasks. Further, to enable real-time inference of the eye signals, we design and develop a mobile edge computing platform, called UbiEi-Edge, which can wirelessly receive the eye signals via low-energy Bluetooth, execute the deep learning algorithm, and visualize decoding results. This real word implementation, developed on an Android Phone, aims to provide real-time data streaming and automatic, real-time decoding of brain visual dynamics, thereby enabling a new paradigm for ALS patients to communicate with the external world. Our experiment has demonstrated the feasibility and effectiveness of the proposed novel mobile edge computing prototype. The study, by innovatively bridging AI, edge computing, and mobile health, will greatly advance the brain dynamics decoding-empowered human-centered computing and smart health big data applications.
Power quality enhancement in transmission system proposes the new structure of Switched Capacitor Multilevel Inverter (SCMLI)-based Static Synchronous Compensator (STATCOM) to improve the power quality. SCMLI can gene...
详细信息
Solar power is a great alternative to existing power sources. Global warming will lead to a lack of power sources soon. With the rapid increase of electrical power in the atmosphere, it is necessary to take strong ste...
详细信息
We present a machine learning approach that uses a custom Convolutional Neural Network (CNN) for estimating the depth of water pools from multispectral drone imagery. Using drones to obtain this information offers a c...
We present a machine learning approach that uses a custom Convolutional Neural Network (CNN) for estimating the depth of water pools from multispectral drone imagery. Using drones to obtain this information offers a cheaper, timely, and more accurate solution compared to alternative methods, such as manual inspection. This information, in turn, represents an asset to identify potential breeding sites of mosquito larvae, which grow only in shallow water pools. As a significant part of the world’s population is affected by mosquito-borne viral infections, including Dengue and Zika, identifying mosquito breeding sites is key to control their spread. Experiments with 5-band drone imagery show that our CNN-based approach is able to measure shallow water depths accurately up to a root mean square error of less than 0.5 cm, outperforming state-of-the-art Random Forest methods and empirical approaches.
Aiming at the problem of how to maintain stable and secure operation of the smart grid after a false injection attack, the tolerant intrusion control of the system after a false data injection attack is investigated. ...
详细信息
A vehicle is a means of transportation, such as a car, truck, or train, that is capable of moving people or goods from one place to another. Vehicles can be classified based on various factors, such as the type of fue...
A vehicle is a means of transportation, such as a car, truck, or train, that is capable of moving people or goods from one place to another. Vehicles can be classified based on various factors, such as the type of fuel they use (e.g. gasoline, diesel, electricity), the number of wheels they have (e.g. two, four, six), and their intended use (e.g. passenger transportation). Vehicles may have connectors, such as plug sockets or fuel ports, that allow them to be connected to other devices or systems to form Vehicle-to-Everything (V2X) technology. For example, an Electric Vehicle (EV) may have a charging port that allows it to be connected to an electric power source to recharge its batteries such Vehicle-to-Grid (V2G) as one of the V2X forms. One of the challenges in charging EVs is the availability of charging infrastructure. In many places, there are relatively few public charging stations, which can make it difficult for EV owners to find a place to charge their vehicles when they are away from home. Additionally, charging an electric vehicle can take significantly longer time than filling up a gasoline-powered vehicle, which can be inconvenient for some drivers. In this review, the various topologies of V2X, connectors, charging challenges, and EV impact types on the grid are conducted.
暂无评论