| 摘要 |
PROBLEM TO BE SOLVED: To provide a sufficiently robust and effective edge detection method for processing a variable condition and an encountering document base body over a wide range. SOLUTION: This edge detection method includes a step (a) for scanning a part of a document for obtaining a set of image data, a step b for calculating a set of first values from the image data by using a first function as a linear function, a step c for calculating a set of second values from the image data by using a second function as a quadratic function, a step d for deciding a first inclination value as a function of a difference between a plurality of first values, a step e for deciding a second inclination value as a function of a difference between a plurality of second values, and a step f for deciding a detection edge of the document from the first and second inclination values.
|
