In this article, we present on original approach for the localization of objects in an image. Our approach is neuronal and it includes two steps. In the first step, a rough localization is performed by presenting each pixel with its neighbourhood to a neural net which is able to indicate if this pixel and its neighbourhood are the image of the search object. This first filter is not very discriminant for the position. From its result, we can select areas which might contain an image of the object. In the second step, these areas are presented to another neural net which can determine the exact position of the object in each area. We apply this algorithm to the problem of localizing faces in images.
|Number of pages||18|
|Journal||Revue technique - Thomson-CSF|
|Publication status||Published - Mar 1993|
ASJC Scopus subject areas