A Fast and Accurate Approach for Computing the Dimensions of Boxes from Single Perspective Images

Leandro A. F. FernandesManuel M. OliveiraRoberto da SilvaGustavo J. Crespo

This paper describes an accurate method for computingthe dimensions of boxes directly from perspectiveprojection images acquired by conventional cameras. Theapproach is based on projective geometry and computesthe box dimensions using data extracted from the box silhouetteand from the projection of two parallel laser beamson one of the imaged faces of the box. In order toidentify the box silhouette, we have developed a statisticalmodel for homogeneous-background-color removalthat works with a moving camera, and an efficient votingscheme for the Hough transform that allows the identificationof almost collinear groups of pixels. We demonstratethe effectiveness of the proposed approach by automaticallycomputing the dimensions of real boxes usinga scanner prototype that implements the algorithms andmethods described in the paper. We also present a discussionof the performed measurements, and an error propagationanalysis that allows the method to estimate, fromeach single video frame, the uncertainty associated to allmeasurements made over that frame, in real-time.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: