In this paper, a tool for describing the geometrical structure of a continuous or discrete, multidimensional signal is considered The origins and foundations of this tool, which we call the pecstrum, lie in the principles of mathematical morphology The pecstrum is defined, its properties are studied, its computation is investigated, and some examples are given The suitability of the pecstrum as a shape descriptor is examined Finally, a practical system for multidimensional shape recognition that uses the pecstrum is introduced, tested, and evaluated

background

Three-dimensional robot vision.

Multidimensional shape description and recognition using mathematical morphology

Vision can extend the use of a robot to a large number of tasks which are beyond that of a “blind” robot. A similar situation applies to sensors operating in simulation of the other human senses and also providing senses not available to the human. This chapter gives an insight into what is possible beyond vision.(1–8)

Three-dimensional robot vision.

Citations

Multidimensional shape description and recognition using mathematical morphology

Nonvision Sensors, Range Finders, and Processing

Related Papers (5)

A combined 2D-3D vision system for automatic robot picking

Robot guidance using computer vision

Research of Positioning of Minimally Invasive Surgery Robot Based on Monocular Vision

3-D object modeling by a camera equipped on a mobile robot

Understanding spatial configuration of robot's environment from stereo motion images