Перегляд за автором "Zhun Shen"
Зараз показано 1 - 1 з 1
Результатів на сторінку
Варіанти сортування
Публікація A WSN Approach to Unmanned Aerial Surveillance of Traffic Anomalies: Some Challenges and Potential Solutions(EWDTS, 2012) David Afolabi; Ka Lok Man; Hai-Ning Liang; Eng Gee Lim; Zhun Shen; Chi-Un Lei; Tomas Krilavičius; Yue Yang; Lixin Cheng; Hahanov, V. I.; Yemelyanov, IgorStationary CCTV cameras are often used to help monitor car movements and detect any anomalies— e.g., accidents, cars going faster than the allowed speed, driving under the influence of alcohol, etc. The height of the cameras can limit their effectiveness and the types of image processing algorithm which can be used. With advancements in the development of inexpensive aerial flying objects and wireless devices, these two technologies can be coupled to support enhanced surveillance. The flying objects can carry multiple cameras and be sent well above the ground to capture and feed video/image information back to a ground station. In addition, because of the height the objects can achieve, they can capture videos and images which could lend themselves more suitably for the application of a variety of video and image processing algorithms to assist analysts in detecting any anomalies. In this paper, we examine some main challenges of using flying objects for surveillance purposes and propose some potential solutions to these challenges. By doing so, we attempt to provide the basis for developing a framework to build a viable system for improved surveillance based on low-cost equipment. With the cost of cars decreasing, more and more people are opting to use cars as their main means of transportation. In cities with large populations, the exponential rise in the number of cars on the streets can lead to many issues (e.g., accidents, congestions, etc.). Governments are spending large amounts of resources in order to improve means to help monitor the movement of cars and in the process enable enforcement officers detect any existing anomalies and prevent potential ones. One widespread technology used to monitor the flow of cars is CCTVs. These can be seen placed on top of street light posts, traffic lights and/or specialized street structures. Although useful, these types of structures are limited in their height, and this limitation can constraint severely the kinds of images and videos can be captured. Similarly, the type of images and videos can determine to a large extent how well they support computer vision and image analysis algorithms. We believe that the use of unmanned flying (or aerial) vehicles (UAV) embedded with video cameras and wireless devices to be used in conjunction with normal CCTVs can support enhanced monitoring of car movements. Unmanned flying objects have become inexpensive and so have video cameras and wireless devices. In this paper, we explore some challenges of using these technologies for automatic monitoring of car flows and suggest some potential solutions for researchers to consider.