Method for separating images and graphics context

Abstract

The present invention relates to a photo / text separation method. It is characterized by that it utilizes the specific intensity and colourity range possessed by every source color in the chromatograph, and converts the file image into several source color planes according to color mode, then executes intensity or colourity operation program for every source color plane so as to can discriminate photo / text type possessed in file image.

Description

technical field [0001] The invention relates to a method related to image processing, in particular to a method applied to image-text separation, using brightness or chroma to separate image-text types in document images. Background technique [0002] With the advent of the digital age, document image analysis technology has been widely used in digital related image processing and applications, such as: license plate recognition, fingerprint recognition, image recognition of military satellite photos, scanners, printers, text recognition software and digital data storage. Whether the file image analysis technology is good or not, and whether the result is correct or not will affect the subsequent processing of the system or whether it can be determined to save more time and space, and even affect the processing capacity of the entire system. Therefore, any document image analysis technology with similar functions will find a method that can correctly and quickly identify do...

AI Technical Summary

Problems solved by technology

[0004] Due to the development of digital information, document images usually contain text and graphic data in solid or mixed colors, and the above-mentioned existing document image recognition technologies all focus on the document image analysis technology for processing grayscale text data. , if the document image has solid-color or mixed-color text and graphic data, but the recognition technology provided can only deal with gray-scale text data, it is actually not very helpful for subsequent processing procedures. For example: use color The printer prints a document image with solid-color text data and mixed-color graphic data. The processing method of the existing document image analysis technology, because the recognition of the edge of the solid-color text data (for example: black) is not very sensitive, so the solid-color text The data is identified as mixed-color graphic data. Therefore, when printing solid-color text data, CMY three-color inks must be used to print, which not only wastes colored inks, but also makes printing difficult due to the need to deal with the mixing of three-color inks. In addition, the effect of the printed text is mixed with colored inks, which is close to pure color instead of pure color, so the printed document image will have color distortion, and the above-mentioned distortion is for some data expressed in color Research material is not allowed at all

[0005] According to the above, due to the existing method of processing document images using the characteristic of grayscale energy concentration, the ability to identify materials with pure colors and mixed colors is not very sensitive, so the existing technology is not suitable for processing files with pure colors and mixed colors. in file images

Images