FRBA - Revista Proyecciones - Vol. 12 Nro. 1

Permanent URI for this collectionhttp://48.217.138.120/handle/20.500.12272/5549

Browse

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Reconocimiento automático de objetos y caracteres en una baraja francesa
    (2014-04-01) Masi, Walter Sebastián; Silvestre, Santiago Germán; Campos, Alfredo Nicolás
    Objects and characters recognition is a common activity on our life. Digital images processing offers high speed automation, objectivity and accuracy. Nowadays there are many objects and characters recognition methodologies. However, none of them shows a solution which could apply for all possible cases. In this paper, pattern comparison, geometric symmetry, Hough transform, color detection and objects count procedures were developed to get the value and suit of cards belonging to a standard French 52-card deck. Different kind and levels of distortions were used, like unfocused, Gaussian noise and rotation distortions. 96% efficiency was obtained as a result of applying all the recognition algorithms in a deck in good condition. In the case of distorted cards, were reached values greater than 75%. The rotated cards (figures mainly) showed the biggest error level.