A Method of Shading Adjustment of Text Image

Abstract

The invention discloses a method for shade adjustment of a character and an image, and relates to the field of character and image processing. The method comprises the steps of marking connected domains of a binary image, carrying out corresponding calculation according to a pixel value inside a minimum enclosing rectangular region and a pixel value of a refining skeleton region of the connected domains, and obtaining a measurement shade value dense of shade of the image. According to the method, firstly, the shade value dense of the character and the image is calculated, the calculated shade value dense is compared with a shade value set by a user, then, a black field value and / or a white field value of the character and the image are / is adjusted, finally, the calculated shade value dense is enabled to meet the shade requirement set by the user, and the shade adjustment of the character and the image is completed. According to the method for the shade adjustment of the character and the image, not only can the thickness of strokes of the character be reflected, but also the shade of the thickness of the strokes of the character can also be reflected, grey-scale maps with various visual effects are obtained, a reference standard is provided for automatic processing of the image, and the method is beneficial for improving efficiency and reducing cost.

Description

Technical field [0001] The invention relates to the field of text image processing, in particular to a method for adjusting the density of text images. Background technique [0002] In text image processing, due to different scanning conditions, the gray distribution of scanned document images is also different. In the actual digital production of layouts, different OCR (Optical Character Recognition, Optical Character Recognition) engines have different accuracy rates for text images with different "shades". Some have a higher recognition rate for "light" text, and some have a higher recognition rate for "dark" text. When the grayscale image is saved as output, the text needs to be adjusted to a level that is comfortable for human eyes to read. [0003] There are many image gray adjustment techniques in the image processing field, such as histogram equalization. However, these general grayscale adjustment techniques are not specifically designed for text scanned images, nor can...

AI Technical Summary

Problems solved by technology

The adjustment method in this patent uses the ratio of black and white pixels in the binary image, but the grayscale image cannot be obtained. Its essence is to adjust the thickness of the text strokes in the binary image

The problem existing in the existing technology is that only using the ratio of the number of black and white pixels in the binary image can only output the binary image, which reflects the thickness of the strokes, but cannot reflect the depth of the grayscale of the text

Images