| 摘要 |
A method captures a video image frame of a book, estimates a position of at least a first endpoint of the book's spine, applies an edge detection operation to the video image frame to generate an edge image, applies a Hough transform to a first region in the edge image to obtain a plurality of line estimates, and rejects line estimates that do not substantially intersect with an estimated endpoint of the book's spine. For line estimates that are not rejected, detecting one or more clusters of angles of the line estimates with respect to an estimated endpoint of the book's spine, and generating an average angle from the cluster of angles. An average angle is selected, and an angular position of the turning leaf in the book's image responsive to the currently selected average angle is estimated.
|
