Binary Image Correction Method Based on Morphology and Grid Construction

Abstract

The example of the present invention provides a binary image correction method based on a self-contained frame, which is applied to digital restoration of paper images. The image correction method includes: scanning the image to be corrected into a grayscale image, and then binarizing it with an empirical value; performing morphological expansion, filtering, and opening operations on the binarized image, and then fitting the above , bottom, left, and right four frame equations; use a set calibration map, print and scan it, and use it to generate a table of vertical and horizontal offset coefficients. The grid is constructed according to the border equation of the image to be corrected and the vertical and horizontal offset coefficient tables, and finally the value of the corresponding pixel after restoration is determined according to the ratio of the number of black and white pixels in each grid. The application of the method can effectively solve the rotation, scaling and non-uniform distortion of the image caused by printing and scanning, the corrected image has little difference from the original image, and the restoration effect is ideal.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image correction method with a built-in frame. Background technique [0002] As the main medium of daily communication, paper materials are often used to carry important information. However, in the copyright certification of paper materials, it is usually necessary to digitally restore the paper materials, and the restoration accuracy is relatively high. [0003] In the process of digital restoration of drawings, it needs to be scanned first, and geometric distortion will inevitably occur in this process, such as rotation, scaling, non-uniform deformation, etc. In the existing image correction methods, most of them require a large number of complex calculations to detect the tilt angle of the image and solve the rotation distortion, such as the method based on Hough transform, the method based on Fourier transform and the method based on cross-correlation. There is...

AI Technical Summary

Problems solved by technology

However, the above methods are difficult to meet the accuracy requirements

Images