CONTROL-FREE LOW-LEVEL IMAGE SEGMENTATION: THEORY, ARCHITECTURE, AND EXPERIMENTATION.

Abstract

The computer vision problem of segmenting images is addressed. The main paradigm of the approach hinges on the fact that low-level image segmentation is a model-driven operation. This model can be conveyed such that all relevant knowledge gathered in a supervised learning phase is used in parallel in the segmentation process. Such a 'control-free' image segmentation can be achieved by using a pattern-recognition approach. This method uses a relatively large number of local image features and combines them optimally according to the scene knowledge acquired in a training phase by the use of a supervised classification procedure. In this methodology, training is performed by the user outlining the image regions belonging to each class. Two main advantages of this approach are that the need of expert image-analysis knowledge is minimized and that it is amenable to parallel pipeline hardware implementation. Experimentation with many different industrial problems demonstrates that this approach is an effective and useful building block for low-level computer vision applications.