| 摘要 |
PROBLEM TO BE SOLVED: To solve the problems that a small amount of characters cannot be distinguished from refuse such as dirt and dust with a histogram and edge information of the entire image when distinguishing whether image data to be read by imaging means is white paper or content and that recycled paper mixed with many impurities cannot be distinguished fro a small amount of characters in the case of using the edge information of the entire image.SOLUTION: Read image data is divided into a plurality of areas, and a histogram is generated and edges are counted in each of the areas. White paper detection is performed on the basis of a dispersion value calculated from the histogram of each area. A variation among the areas is calculated from an edge number of each area, and white paper detection is performed according to the magnitude of the variation. Whether to be white paper or content is determined on the basis of a detection result due to the histogram and a detection result of the edge number. |
