Complex background text image extraction method and system based on multi-threshold fusion

Abstract

The invention discloses a complex background text image extraction method and system based on multi-threshold fusion. The method comprises the following steps: S1, acquiring an original image, and acquiring a local binarization parameter according to image category information; S2, converting the original image into a grayscale image; S3, obtaining an edge-based first binary image of the grayscaleimage by using a first processing algorithm; performing local binarization processing on the grayscale image based on the local binarization parameter by using a second processing algorithm to obtaina second binary image; and S4, carrying out logic OR operation on the first binary image and the second binary image, outputting the two images, and extracting the text image. The first binary imagebased on the edge and a second binary image obtained by local binarization processing are fused, a defect that foreground characters are lost due to noise removal in the second binary image is effectively overcome, and the local binarization parameters are set according to the category of the to-be-processed image, so that the processing effect on the characters with shallow strokes is very excellent, and the quality of the binarized image is improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method and system for extracting complex background text images based on multi-threshold fusion. Background technique [0002] In our daily life, countless copies of paper documents are produced every day, and in many cases, our paper documents need to be electronically scanned and stored as electronic files. On the one hand, due to the poor storage conditions of some paper documents, the documents may be stained or creased, so that the quality of the scanned electronic documents is not high; On the other hand, if you encounter a paper document written on both sides, when scanning, the text on the back will also appear on the front; due to the above aspects, the text image has stains, creases, etc. , uneven lighting, skewed documents, and text reflections on the back (complex background), which become complex background text images, such as figure 2 As shown, the imag...

AI Technical Summary

Problems solved by technology

With the uncertain pixel as the center, iterate to the surrounding local area. When the sum of the foreground pixels in the local area is greater than the sum of the background pixels, it is judged as the foreground pixel, otherwise it is the background pixel. When the sum is the same, the local area is expanded for judgment. This The method is based on a clustering assumption that pixels of the same category should be closer together. Although the document image can reach the same size after normalization, but the characters inside are different at this time, it cannot be consistent. standard, so the above binarization method cannot be used simply for binarization, and the effect is not good

Images