High dynamic range acquisition and shape recognition

The paper presents different approaches in improving the digital image acquisition with focus on shape recognition. The study is part of a larger project focused on image pre-processing optimization in order to obtain a higher efficiency of automated features recognition. The final target would be to use the results in the field of number plate recognition in road traffic. Structured in two main sections, the work started with several small projects targeted at improving the image acquisition in the process of welding observation. Emphasis is being set on presenting an introduction to the state of the art in shape recognition, and the ultimate intention is the one of improving such techniques. We intend to apply afterwards the results on the practical application mentioned before, namely the number plates recognition in road traffic. Noticeable results have been obtained in the area of improving the high dynamic range at image acquisition. Several techniques have been tested in order to obtain a better dynamic range in the field of welding observation.