TWI256836B - Method for correcting the image frame-selection range on image input device - Google Patents

Abstract

The present invention provides a method for correcting the image frame-selection range on image input device, which employs four corner data for the image obtained by scanning from an image input device. When the scanned image is a rectangular block, the method can operate to obtain the most possible four correct corners for the scanned image, so as to precisely determine the frame-selection range. The method includes the following steps: (A) loading the coordinates for four corners of the scanned image; (B) determining whether the quadrilateral connected with the four corners of the scanned image is a rectangle, if so, proceeding to the step (E), otherwise, proceeding to the step (C); (C) determining whether the quadrilateral has at least one vertical internal angle, if so, proceeding to the step (D), otherwise terminating the image frame-selection procedure; (D) using the two sides of the vertical internal angle to generate vertical lines from the ends of the segment of the two sides, taking the intersection of the two vertical lines as one new corner of the scanned image, and taking the four intersections of the two sides of the vertical internal corners and the two vertical lines as the corners for the scanned image; and (E) outputting the coordinates of four corners for the scanned image after confirmation or adjustment.