An object-oriented language for image and vision execution (OLIVE)

The authors present OLIVE, an object-oriented language for machine vision and image processing that is intended to make it easier to develop efficient, portable applications. The goal in defining OLIVE is to provide a well-factored, intuitive, portable tool for image processing and analysis that insulates the user from details of the implementation while at the same time exploiting its capabilities (e.g. specialized hardware). The authors define OLIVE's principal object types-images and loci (abstractions of point sets and geometric entities)-and their corresponding operations, including the use of loci as generalized indexes for images. They present several examples of OLIVE for typical image processing and machine vision tasks. They discuss issues in the implementation of OLIVE, including a hardware architecture that simplifies the implementation while enhancing its performance.<<ETX>>