Method for segmenting and extracting name card area image under noise pollution

Abstract

The invention discloses a method for segmenting and extracting a name card area image under noise pollution. The method comprises the steps that (1) color channel transformation is conducted, wherein a color image is transformed into a grey-scale image; (2) noise is suppressed through multi-cycle bilateral filter; (3) the background is removed through seed growth; (4) flaws and burrs are removed through morphology correction. According to the method, for the name card image, a name card area is rapidly segmented and extracted through the design of combining related parameters and by taking grey-scale transformation, bilateral filter noise suppression, seed growth, morphology correction and the like into consideration. In the method, a good segmentation result can be rapidly obtained simply by inputting the name card image shot by a mobile phone, and the name card area can be directly obtained through the segmentation result. The method can be applied to name card extracting and processing through mobile phone images, camera shot images and the like, and the final recognition of the name cards is assisted.

Description

technical field [0001] The invention relates to an image processing technology, in particular to an image segmentation and extraction method of a business card area under noise pollution. Background technique [0002] In contemporary society, as a carrier of effective information, personal business cards are used more and more frequently. Therefore, although the business card is small, it has a close relationship with business opportunities and benefits. This shows that the collection and arrangement of business card information is very important. However, there are more and more business cards, and the requirements for information management are getting higher and higher. The method of manually carrying paper is no longer applicable. In addition, the method of manually entering mobile phones and other electronic devices is very time-consuming and labor-intensive. . This has led to the great development of the business card recognition work of mobile phone programs at thi...

AI Technical Summary

Problems solved by technology

First of all, the images taken by mobile phones are random, which mainly refers to the relationship between illuminance and reflectivity, which leads to the image noise and uneven illuminance. The grayscale images obtained on this basis also contain a lot of noise. Here Basic binarization will encounter great challenges, even if morphological denoising is used later, it will not be able to solve the problem of losing more information in binarization; secondly, usually captured images, including business cards, will also contain Part of the background, maybe the desktop, maybe the subsurface of the notebook, etc., will interfere with the division of the business card area, and we must consider this issue

Images