Basically template match with rotation. Given a template as binary edge image the function find the object that match the template in the image. This work even if the template and object are in different orientations. The matching is is done by rotating the template and matching it to the image in various of angles, the best match is returned. Matching the template to the image can be done by various of template match approaches including crosscorrelation and generalized hough transform. See the readme file supplied in the code zip file for more details