Classificação automática de padrões projetados por um sistema de luz estruturada

Abstract

One of the main problems in Computer Vision and Close Range Digital Photogrammetry is 3D reconstruction. 3D reconstruction with structured light is one of the existing techniques and which still has several problems, one of them the identification or classification of the projected targets. Approaching this problem is the goal of this paper. An area based method called template matching was used for target classification. This method performs detection of area similarity by correlation, which measures the similarity between the reference and search windows, using a suitable correlation function. In this paper the modified cross covariance function was used, which presented the best results. A strategy was developed for adaptative resampling of the patterns, which solved the problem of deformation of the targets due to object surface inclination. Experiments with simulated and real data were performed in order to assess the efficiency of the proposed methodology for target detection. The results showed that the proposed classification strategy works properly, identifying 98% of targets in plane surfaces and 93% in oblique surfaces.