Parallel Architectures and Computer Vision Tapa dura - 1988

Información de la editorial

The computer interpretation of visual images offers unlimited potential, with applications ranging from robotics and manufacturing to electronic sensors for aiding the blind. However, there is a huge gap between the promise of technology and what is actually possible now. In order to work
effectively, computers will have to sense and analyze visual scenes in a fraction of a second, but currently it is not unusual to devote an hour of computer time to the analysis of a single image. Also, such images often have to be of highly stylized scenes to make any analysis possible. The only
hope for the future lies in the use of massive parallel architectures, with perhaps thousands of processors cooperating on the task. Fortunately, the spectacular advances now being made in VLSI technology may allow such parallelism to be economically feasible. This book draws together the
proceedings of a key workshop held in 1987. It presents the work of leading U.K. researchers in parallel architectures and computer vision from both industry and academia, providing a clear indication of the state of the art.