Method and device for image data deflection detection and correction

Abstract

The invention relates to a method and a device for image data deflection detection and correction, which are used for image data and image processing devices. The device comprises a deflection detecting module and an image rotation correcting module. The method comprises the following steps: the deflection detecting module receives image data and carries out binary image processing to the image data so as to obtain binary image data; the binary image data is rotated in multiple angles to obtain a plurality of binary image data to be detected; the binary image data to be detected is processed to obtain a plurality of horizontal projection quantity calculating curves; the horizontal projection quantity calculating curves are respectively processed by a high-pass filtering program to obtain a plurality of high-frequency signal curves, and a plurality of corresponding index values are obtained by calculation; a deflection angle of the image data is obtained by detection according to the maximum value in the index values; and the image rotation correcting module is used for rotation correction for the image data according to the deflection angle, and then the corrected image data is obtained.

Description

technical field [0001] The present invention is a method and device for skew detection and correction of image data, especially the method and device for skew detection and correction of image data used in image processing devices. Background technique [0002] With the improvement of photoelectric technology, the industry has developed many devices that convert paper documents into electromagnetic data files, such as: scanners, copiers, fax machines or physical projectors, etc., in order to eliminate the need for paper documents to be scanned or photographed. Inadvertently placing the document skewed, or improving the document skew that may be caused during the paper transmission process, some document image skew detection and correction (skewed document detection / correction) technologies are used in various devices. However, most of the difficulties of this type of technology exist in how to remove the non-text graphics in the document. Wherein the more critical existing ...

AI Technical Summary

Problems solved by technology

In addition to removing the graphics in the file, there are also people who ignore the file content and use the difference between the file color and the background color to get the four corners or borders of the file, and directly get the amount of skew. However, this technology has little difference in color or If the boundary is not within the scanning range, it will not take effect

But in any case, most of the existing reliable estimation methods are inseparable from the need for a large amount of memory to store classified and statistical information, which is not conducive to the use of hardware.

Images