Argus: A flexible real-time system for 2d defect and texture classification of wooden material

This paper describes a system for real-time inspection of 2D surfaces. It was initially planned as system for classification of wooden surfaces, but was successfully used also in the context of other inspection tasks like metallic surface inspection and leather inspection. The system has two major modules. One is a 2D object segmentation and recognition part, where key elements of the underlying elements have been published before. This includes hierarchical processing of the incoming gray-level images leading to a symbolic description of the surface; syntactic segmentation; and the decision network methodology used. Beyond these features, a new track has been added, which is
 entirely devoted to texture classification in real-time. This two-way analysis of wooden surfaces was first implemented on a heterogeneous architecture containing Zoran vector processors and Transputers (all commercially available). The current version uses only TMS32C40 processors. The system has been successfully implemented in a production plant in Austria. We describe major elements of the system and the underlying algorithms.