Automatic recognition of individuals is a significant problem in the development of pattern recognition. In this paper, we introduce a simple technique for personal identification of human faces in cluttered scenes ba...
详细信息
Automatic recognition of individuals is a significant problem in the development of pattern recognition. In this paper, we introduce a simple technique for personal identification of human faces in cluttered scenes based on neural nets. In the detection phase, neural nets are used to test whether a window of 20/spl times/20 pixels contains a face or not. A major difficulty in the learning process comes from the large database required for face/nonface images. We solve this problem by dividing the data into two groups. Such division results in reduction of computational complexity, thus decreasing the time and memory needed during the image test. For the recognition phase, feature measurements are made through Fourier descriptors. Such features are modified to reduce the number of neurons in the hidden layer. Simulation results for the proposed algorithm show good performance compared with previous results.
As telecomm.nications service features become more plentiful, service providers are faced with the increasingly difficult task of providing consistent and high-quality service. A stumbling block along the path towards...
详细信息
As telecomm.nications service features become more plentiful, service providers are faced with the increasingly difficult task of providing consistent and high-quality service. A stumbling block along the path towards making available large numbers of complex and potentially highly customized features are the unwanted or unforeseen behaviour of these features when operating together. This problem, known as the feature interaction problem, is only compounded by the heterogeneity of service providers, service platforms, and feature implementations. Presented in this paper is a constraint-based approach for performing avoidance, detection, and resolution of these feature interactions in a run-time telecomm.nications environment, such as an intelligent network. This constraint representation centres around a framework for call resources, permitting structured access to each of the aspects of call functionality, allowing features to remain modular. The representation is further enhanced, beyond the basic finite state feature processes, by the maintenance of context information. The resulting call session information keeps track of information such as how a feature is being used, by whom it was invoked and when, and what other callers and features are currently involved.
Colorimetric methodology using enzyme activity as a quantitative label has been applied to a variety of easy, standardised and highly sensitive tests. In the p24 antigen assay - the only inexpensive quantitative test ...
详细信息
作者:
BOHM, SELHAKEEM, AKHACHICHA, MDepartment of Electrical and Computer Engineering
Concordia University 1455 De Maisonneuve Blvd. West Montreal H3A 1M8 Canada Was born in Montreal
Canada on 14 September 1966. He received the B. Eng. degree in electrical engineering from Concordia University Montreal Canada in 1989. He is at present completing the M.A.Sc. degree in electrical engineering at Concordia University. (S'75–S'79–M'79–SM'86) received the Ph.D. degree from Southern Methodist University
Dallas TX in 1979. He spent the next two years working as a Visiting Professor in Egypt after which he moved to Ottawa Canada in 1982. He assumed teaching and research positions in Carleton and Manitoba Univerities and later moved to Concordia University Montreal Canada in 1983 where he is now a Professor in the Electrical and Computer Engineering Department. He has published numerous papers in IEEE and international journals in the areas of spread spectrum and networking. He is a well-known expert in these areas and serves as a consultant to many companies. His current research interests include wide-band metropolitan networks switching architectures and performance of on-board multibeam satellites acquisitionless CDMA networks code distribution and orthogonalization of CDMA signals responsive congestion control for ATM-based networks ARQ techniques and investigation of the novel SUGAR CDMA systems in fading channels. Dr. Elhakeem is a Senior Member of the Canadian Electrical Engineering Society and Armed Forces Association. He has chaired numerous technical sessions in IEEE Conferences was the Technical Program Chairman for IEEE Montech 1986 Montreal Canada. Dr. Elhakeem is the key guest editor of theIEEE Journal of Selected Areas in Communicationsfor the May June issues 1993 covering CDMA networks. Advanced Technology & Networks
VISTAR Telecommunications Inc. Ottawa Ontario K1G 3J4 Canada An Associate Director of Advanced Technology & Networks Group
VISTAR Telecommunications Inc. Ottawa Canada. He is also an Adjunct Pr
In this paper, we study the performance of a prioritized on-board baseband switch in conjunction with a multibeam satellite handling integrated services. The services considered for the analysis include voice, video, ...
详细信息
In this paper, we study the performance of a prioritized on-board baseband switch in conjunction with a multibeam satellite handling integrated services. The services considered for the analysis include voice, video, file transfer and interactive data. The prioritized switch uses both input and output buffering, switch speed-up as well as a two-phase head-of-line resolution algorithm, in order to reduce the buffer loss while maintaining acceptable user delays. The minimum required buffer capacity and switch speed-up for each service in a prioritized environment are found under uniform traffic conditions. It is shown that under uniform traffic conditions, only minimal buffering and switch speed-up are needed even for the lowest priority users. The performance dependence on the switch size is also substantially reduced with head of line resolution and buffering even in a prioritized environment.
暂无评论