This work focuses on the recognition of three-dimensional colon polyps captured by an active stereo vision sensor. The detection algorithm consists of SVM classifier trained on robust feature descriptors. The study is related to Cyclope, this prototype sensor allows real time 3D object reconstruction and continues to be optimized technically to improve its classification task by differentiation between hyperplastic and adenomatous polyps. Experimental results were encouraging and show correct classification rate of approximately 97%. The work contains detailed statistics about the detection rate and the computing complexity. Inspired by intensity histogram, the work shows a new approach that extracts a set of features based on depth histogram and combines stereo measurement with SVM classifiers to correctly classify benign and malignant polyps.

International audienceIn this paper we focus on the recognition of threedimensional objects captured by an active stereo vision sensor. The study is related to our research project Cyclope, this embedded sensor based on active stereo-vision approach allows real time 3D objects reconstruction. Our medical application requires differentiation between hyperplastic and adenomatous polyps during 3D endoscopic imaging. The detection algorithm consists of SVM classifier trained on robust feature descriptors of a surfacic 3D point cloud extracted from the surface of studied object. We compared our feature extraction method with others. Experimental results were encouraging and show correct classification rate of approximately 97%. The work contains many techniques concerning image processing and system calibration and provides detailed statistics about the detection rate and the computing complexity