In this work, we present the parallel implementation of three low-Ievel vision algorithms, namely smoothing, histogram generation and edge detection by using the Sobel operator on the Homogeneous Multiprocessor. These algorithms were run on the simulator specifically developed for the Homogeneous Multiprocessor and the simulation experiments were used to establish the performance of these algor...

This paper analyses the requirements of a toolset intended for integrated parallel system design Real time data dominated systems are targeted particularly those in vision and image processing A con strained abstract model is represented based around software pipelines each stage of which can contain a parallel component The toolset is novel in seeking to enable generalized parallelism in which...

|This paper describes a method for vehicle recognition, in particular, for recognizing a vehicle's make and model. Our system is designed to take into account the fact that vehicles of the same make and model number come in di erent colors; to deal with this problem, our system employs infra-red images, thereby eliminating color di erences. Another reason for the use of infra-red images is that...

* This work has been supported by the Spanish CICYT under grants TIC97-0897-C04-03 and TIC98-0559. Abstract In this paper we propose several applications of the EM algorithm –a well-known algorithm for parameter estimation in generic statistical problems– in the different levels of a computer vision system, and discuss its relatively straightforward and efficient implementation in parallel dist...

Vision chips are microelectronic devices which combine image sensing and processing on a single silicon die. In a way somewhat resembling the vertebrate retina these VLSI chips perform preliminary image processing directly on the sensory plane and are capable of very high processing speed at very low power consumption. This makes them particularly suitable for applications such as autonomous ro...

111 this work a nlassively parallel VLSI architectu re is discussed: PAPRICA, a mesh connected machine that can also simulate a pyramidal architecture. Its computational paradigm is based on the lnatchiug operator as defined in nlathematical morphology. This description is equivalent to the Cellular Automata paradigm, which can simplify the design of PAPRICA applications in low-level vision. PA...

This paper proposes a novel technique to computing geometric information from images captured under parallel projections. Parallel images are desirable for stereo reconstruction because parallel projection significantly reduces foreshortening. As a result, correlation based matching becomes more effective. Since parallel projection cameras are not commonly available, we construct parallel image...

The construction of a parallel vision system from Standard ML prototypes is presented. The system recognises 3D objects from 2D scenes through edge detection, grouping of edges into straight lines and line junction based model matching. Functional prototyping for parallelism is illustrated through the development of the straight line detection component. The assemblage of the whole system from ...

Parallel implementations of two computer vision algorithms on distributed cluster platforms are described. The rst algorithm is a square-error data clustering method whose parallel implementation is based on the well-known sequential CLUSTER program. The second algorithm is a motion parameter estimation algorithm used to determine correspondence between two images taken of the same scene. Both ...