Image processing method and image processing device

Abstract

The invention relates to an image processing method and an image processing device. The method is used for receiving an input image and separating pixels with the character characteristic and pixels with the graph characteristic in the input image. The image processing method comprises the following steps of: performing primary filtering on the input image to obtain a first image processing result; performing secondary filtering on the first image processing result to obtain a second image processing result, wherein the filtering coefficient distribution of the primary filtering is different from that of the secondary filtering; obtaining a group of reference values according to the first image processing result and the second image processing result; and judging whether each pixel in theinput image is a character pixel or a graph pixel according to at least the group of reference values and a predetermined threshold value. By the method and the device, the definition of characters in the input image can be kept, the influence of halftone and other noise on a graph part in the input image can be reduced, and the visual effect of the image is improved.

Description

technical field [0001] The invention relates to an image processing technology and a related device, in particular to an image processing method and an image processing device capable of separating characters and graphics in an input image. Background technique [0002] After the scanner extracts the image data, if it can detect the text part in the image data, print the text part with pure black ink and make different printing settings for the non-text graphics part, not only can Improving the printing quality and efficiency can greatly improve the readability and clarity of the image data. In addition to using edge detection to cut out the text part, the known image-text separation technology also uses methods such as chroma detection, dot detection, and complex-structured filters to exclude non-text graphic parts. However, due to the need to go through With a large number of hardware devices such as calculations and filters, the calculation cost of known edge detection t...

AI Technical Summary

Problems solved by technology

In addition to using edge detection to cut out the text part, the known image-text separation technology also uses methods such as chroma detection, dot detection, and complex-structured filters to exclude non-text graphic parts. However, due to the need to go through With a large number of hardware devices such as calculations and filters, the calculation cost of known edge detection technology is often too expensive. Therefore, how to quickly and efficiently separate text and graphics in an input image is still a challenge in this field. a big subject

[0003] In addition, due to the limitation of printing hardware equipment, half-tone (half-tone) processing, insufficient resolution, errors and noises caused by digitalization often lead to poor image quality and cloud patterns (moire pattern), insufficient text edge sharpness, insufficient dynamic range, and excessive output noise

Images