Traffic Characterization, Application Identification, Per Application Classification, and VPN/Non-VPN Traffic Characterization have been some of the most notable research topics over the past few years. Deep Packet Inspection (DPI) promises an increase in Quality of Service (QoS) for Internet Service Providers (ISPs), simplifies network management and plays a vital role in content censoring. DPI has been used to help ease the flow of network traffic. For instance, if there is a high priority message, DPI could be used to enable high-priority information to pass through immediately, ahead of

Software issues related to automotive controls account for an increasingly large percentage of the overall vehicles recalled. To alleviate this problem, vehicle diagnosis and maintenance systems are increasingly being performed remotely, that is while the vehicle is being driven without need for factory recall and there is strong consumer interest in Remote Diagnosis and Maintenance (RD&M) systems. Such systems are developed with different building blocks/elements and various capabilities. This paper presents a novel automotive RD&M system and prognosis architecture. The elements of the

In vision-based systems, cast shadow detection is one of the key problems that must be alleviated in order to achieve robust segmentation of moving objects. Most methods for shadow detection require significant human input and they work in static settings. This paper proposes a novel approach for adaptive shadow detection by using semi-supervised learning which is a technique that has been widely utilized in various pattern recognition applications and exploits the use of labeled and unlabeled data to improve classification. The approach can be summarized as follows: First, we extract color

Human action recognition is an important area of research in computer vision. Its applications include surveillance systems, patient monitoring, human-computer interaction, just to name a few. Numerous techniques have been developed to solve this problem in 2D and 3D spaces. However most of the existing techniques are view-dependent. In this paper we propose a novel view-independent action recognition algorithm based on the motion history of skeletons in 3D. First, we compute a skeleton for each volume and a motion history for each action. Then, alignment is performed using cylindrical

Most commercial Small Unmanned Aerial Vehicles (SUAVs) rely solely on Global Navigation Satellite Systems (GNSSs) - such as GPS and GLONASS - to perform localization tasks during the execution of autonomous navigation activities. Despite being fast and accurate, satellite-based navigation systems have typical vulnerabilities and pitfalls in urban settings that may prevent successful drone localization. This paper presents the novel concept of 'Deep Urban Signatures' where a deep convolutional neural network is used to compute a unique characterization for each urban area or district based on

In this paper we present a novel framework for geolocalizing Unmanned Aerial Vehicles (UAVs) using only their onboard camera. The framework exploits the abundance of satellite imagery, along with established computer vision and deep learning methods, to locate the UAV in a satellite imagery map. It utilizes the contextual information extracted from the scene to attain increased geolocalization accuracy and enable navigation without the use of a Global Positioning System (GPS), which is advantageous in GPS-denied environments and provides additional enhancement to existing GPS-based systems

Semantic segmentation of 3D point cloud data is essential for enhanced high-level perception in autonomous platforms. Furthermore, given the increasing deployment of LiDAR sensors onboard of cars and drones, a special emphasis is also placed on non-computationally intensive algorithms that operate on mobile GPUs. Previous efficient state-of-the-art methods relied on 2D spherical projection of point clouds as input for 2D fully convolutional neural networks to balance the accuracy-speed trade-off. This paper introduces a novel approach for 3D point cloud semantic segmentation that exploits

In this paper, we describe a method for Scale-Adaptive visual tracking using compressive sensing. Instead of using scale-invariant-features to estimate the object size every few frames, we use the compressed features at different scale then perform a second stage of classification to detect the best-fit scale. We describe the proposed mechanism of how we implement the Bayesian Classifier used in the algorithm and how to tune the classifier to address the scaling problem and the method of selecting the positive training samples and negative training samples of different scales. The obtained

Human motion analysis is a vital research area for healthcare systems. The increasing need for automated activity analysis inspired the design of low cost wireless sensors that can capture information under free living conditions. Body and Visual Sensor Networks can easily record human behavior within a home environment. In this paper we propose a multiple classifier system that uses time series data for human motion analysis. The proposed approach adaptively integrates feature extraction and distance based techniques for classifying impaired and normal walking gaits. Information from body

Human motion analysis provides a valuable solution for monitoring the wellbeing of the elderly, quantifying post-operative patient recovery and monitoring the progression of neurodegenerative diseases such as Parkinson's. The development of accurate motion analysis models, however, requires the integration of multi-sensing modalities and the utilization of appropriate data analysis techniques. This paper describes a robust framework for improved patient motion analysis by integrating information captured by body and visual sensor networks. Real-time target extraction is applied and a