This study aims to develop a method for automating the initial step (performed manually) in the identiï¬cation of the cutting window, on a cloth cuttingequipment. The proposed approach employs computer vision techniques based on two main algorithms, Canny (creating contour map) and Hough (identiï¬cation of lines). Recently the prototype implementation is ï¬nished, providing the availability of the method for evaluation. The next stage of work consists in acquire a set of images as well as the creation of a database with annotation performed by human observers. With the completion of this data base images enable a perform quantitative tests in order to evaluate the accuracy of the proposed method.