Fuer Laurent Letellier 2 Eintraege in Db. L. Letellier und Laurent Letellier. Sind identisch. A Way to Exploit Highlights in Metallic Environment for Pipes Localization

Reprocessing cells of nuclear plants are constituted by a great number of specular metallic pipes. When they are illuminated, this kind of pipes generates areas of high light intensity, called highlights. The latter can produce erroneous results, especially in edge detection. The matter of this paper is to suppress and further to exploit the highlight effects. With this aim in view, we propose a system based on the polarization of light, which not only suppresses highlights in the scene but also uses them to extract relevant information like the pipes axes at the video rate. Fig~we 1 : 1)atahase and real scene. For example, in edge detection, highlights bring virtual Interventions in reprocessing cells of nuclear plants are edges which can be taken for the real pipe edges (Fig. 2). often performed by teleoperated robots equipped with a camera and a light source. It is difficult for the operator to steer itself in this kind of cell, because it is constituted of a great number of repetitive structures of pipes. Our work aims at providing a system able to localize the robot inside the cell at any time to help the operator. A database of the cell is provided (Fig. I), it may be incomplete-some pipes may be missing in the database, or on the contrary some pipes of the database may not exist in the real cell-or imperfect-the position or the orientation of some pipes in the database may differ from these of the corresponding pipes in the real cell-. The detection of the pipes is a crucial part of all the treatment. So, we were first interested in providing an as much performing as possible low level part. In addition, the pipes in the cell are metallic. Thus, the pipe illumination by a light source brings areas of high light intensity, called highlights. Highlights often cause image processing failures, which lead to image misinterpretation .