| 摘要 |
<p>The present invention is directed to watermarking an image using the highest practical watermark intensity that can be used without creating visual artifacts. With the present invention, an image is divided into areas. The size and shape of each area are determined so that each area only contains portions of the image where it is appropriate to use a watermark signal having the same energy (A). Some images consist of a series of lines. Such images are often used as the background image on security documents or currency (A). If an image consists of a series of lines, the invention can be applied by dividing the image into areas where the lines that form the image are within a certain range of widths (Tile). An appropriate watermark intensity is then determined for each such area and a watermark of this intensity is applied to the image. If an image is a grayscale image, areas having the same tonal value or tonal density are determined directly, or the image is transformed into a line art image and areas with the same size binary valued areas (lines, dots, squares, etc.) are selected by filtering the image. After the image has been divided areas which have a line width or tonal density value that is within a certain range of values, an appropriate watermark energy level is assigned to each area. The image is watermarked by changing the value of each bit or pixel in the image by an amount determined by both the value a payload tile and the intensity value associated with the area in which the bit or pixel lies.</p> |
