Vision-based mobile robot navigation is a vibrant area of research with numerous algorithms having been developed, the vast majority which either belong to scene-oriented simultaneous localization and mapping (SLAM) or fall into category robot-oriented lane-detection/trajectory tracking. These methods suffer from high computational cost require stringent labelling calibration efforts. To address these challenges, this paper proposes lightweight framework based purely on uncalibrated...

The dynamic environment due to traffic mobility and wireless communication from/to vehicles make identity authentication trust management for privacy preservation based on vehicular edge computing (VEC) an increasingly important problem in networks. However, existing schemes mainly focus between a single trusted node multiple vehicles. This framework may suffer the bottleneck node, performance depends heavily its resources. In this paper, blockchain empowered group-authentication scheme is...

With the popularization and wide application of drones in military civilian fields, safety must be considered. At present, failure drop rates are still much higher than those manned aircraft. Therefore, it is imperative to improve research on safe landing recovery drones. However, most drone navigation methods rely global positioning system (GPS) signals. When GPS signals missing, these cannot land or recover properly. In fact, with help optical equipment image recognition technology,...

Vehicle surveillance of a wide area allows us to learn much about the daily activities and traffic information. With rapid development remote sensing, satellite video has become an important data source for vehicle detection, which provides broader field surveillance. The achieved work generally focuses on aerial with moderately-sized objects based feature extraction. However, moving vehicles in imagery range from just few pixels dozens exhibit low contrast respect background, makes it hard...

Along with an integration of wearable devices, wireless communications and big data in the smart health, biomedical is collected referring to multiple associated patients during interactions. Due communication channel openness sensibility, privacy preservation become increasingly noteworthy edge cloud hybrid computing-based healthcare applications. In this paper, a cooperative scheme designed for devices identity authentication access control considerations space-aware time-aware contexts....

With the rapid development of Unmanned Aerial Vehicle (UAV) systems, autonomous landing a UAV on moving Ground (UGV) has received extensive attention as key technology. At present, this technology is confronted with such problems operating in GPS-denied environments, low accuracy target location, poor precision relative motion estimation, delayed control responses, slow processing speeds, and stability. To address these issues, we present hybrid camera array-based that can land UGV...

Infrared human action recognition has many advantages, i.e., it is insensitive to illumination change, appearance variability, and shadows. Existing methods for infrared are either based on spatial or local temporal information, however, the global which can better describe movements of body parts across whole video, not considered. In this letter, we propose a novel representation named optical-flow stacked difference image (OFSDI) extract robust discriminative feature from data by...

An infrared sensor is a commonly used imaging device. Unmanned aerial vehicles, the most promising moving platform, each play vital role in their own field, respectively. However, two devices are seldom combined automatic ground vehicle detection tasks. Therefore, how to make full use of them—especially based on imagery–has aroused wide academic concern. due imagery’s low-resolution and detection’s complexity, extract remarkable features handle pose variations, view changes as well...

Recognizing human actions from varied views is challenging due to huge appearance variations in different views. The key this problem learn discriminant view-invariant representations generalizing well across In paper, we address by learning hierarchically using a novel method, referred as joint sparse representation and distribution adaptation. To obtain robust informative feature representations, first incorporate sample-affinity matrix into the marginalized Stacked Denoising Autoencoder...

This paper proposes a novel infrared camera array guidance system with capability to track and provide real time position speed of fixed-wing Unmanned air vehicle (UAV) during landing process. The mainly include three parts: (1) Infrared near laser lamp based cooperative long range optical imaging module; (2) Large scale outdoor calibration (3) Laser marker detection 3D tracking module. Extensive automatic experiments flight demonstrate that our has the unique ability guide UAV safely...

With the rapid development of unmanned aerial vehicles (UAVs), UAV-based intelligent airborne surveillance systems represented by real-time ground vehicle speed estimation have attracted wide attention from researchers. However, there are still many challenges in extracting information UAV videos, including dynamic moving background, small target size, complicated environment, and diverse scenes. In this paper, we propose a novel adaptive framework for multi-vehicle videos. Firstly, build...

DOTS (Dynamic Object Tracking System) is an indoor, real-time, multi-camera surveillance system, deployed in a real office setting. combines video analysis and user interface components to enable security personnel effectively monitor views of interest perform tasks such as tracking person. The component performs feature-level foreground segmentation with reliable results even under complex conditions. It incorporates efficient greedy-search approach for multiple people through occlusion...

In recent years, unmanned aerial vehicles (UAVs) have rapidly developed, but the illegal use of UAVs by civilians has resulted in disorder and security risks increasingly triggered community concern worry. Therefore, monitoring recycling key regions are great significance. This paper presents novel panoramic UAV surveillance autonomous system that is based on unique structure-free fisheye camera array capability real-time detection, 3D localization, tracking, capacity over a very wide field...

Fast and accurate segmentation of moving objects in video sequences is a basic task many computer vision analysis applications. It has critical impact on the performance object tracking classification activity analysis. This paper presents effective methods for solving this problem. Firstly, fast efficient algorithm presented background update to handle various sources scene changes, including ghosts, left objects, camera shaking, abrupt illumination changes. done by analyzing properties...

We describe a new interaction technique that allows users to control nonlinear video playback by directly manipulating objects seen in the video. This is similar "scrubbing" where user adjusts time moving mouse along slider. Our approach superior variable-scale scrubbing can concentrate on interesting and does not have guess how long will stay view. method relies tracking system tracks fixed cameras, maps them into 3D space, handles hand-offs between cameras. In addition dragging visible...

Passive Keyless Entry and Start (PKES) systems are popularly embed in modern cars, which allow users to open start their cars while having car keys' pockets. They bring convenience but vulnerable relay attacks. A attack PKES is a widely known against the challenge- response technique used passive keyless vehicle system, allows true distance between key remained large. The main countermeasure attacks use of bounding protocols measuring round-trip time key. However, most schemes tend more...

Robust detection and tracking of multiple people in cluttered crowded scenes with severe occlusion is a significant challenge for many computer vision applications. In this paper, we present novel hybrid synthetic aperture imaging model to solve problem. The main characteristics approach are as follows. 1) To the best our knowledge, first attempt occluded problem joint camera domain. 2) A framework designed achieve seamless interaction among detection, modules. 3) object module,...

Reasonable house price prediction is a meaningful task, and the clustering an important process in prediction. In this paper, we propose method of Multi-Scale Affinity Propagation(MSAP) aggregating appropriately by landmark facility. Then each cluster, using Linear Regression model with Normal Noise(LRNN) predicts reasonable price, which verified increasing number renting reviews. Experiments show that precision improved greatly via MSAP.

Limited by aircraft flight altitude and camera parameters, it is necessary to obtain wide-angle panoramas quickly stitching aerial images, which helpful in rapid disaster investigation, recovery after earthquakes, reconnaissance. However, most existing algorithms do not simultaneously meet practical real-time, robustness, accuracy requirements, especially the case of a long-distance multistrip flight. In this paper, we propose novel image-only real-time UAV image mosaic framework for flights...

Geolocating multiple targets of interest on the ground from an aerial platform is important activity in many applications, such as visual surveillance. However, due to limited measurement accuracy commonly used airborne sensors (including altimeters, accelerometer, gyroscopes, and so on) small size, complex motion, a large number images, most current unmanned vehicle (UAV)-based target geolocation algorithms have difficulty obtaining accurate geographic location coordinates online,...