Fast document type recognition method based on full-sized feature extraction

Abstract

The invention provides a fast document type recognition method based on full-sized feature extraction. The method comprises the following steps of document image preprocessing, including image zooming, graying and noise filtering; document image feature extraction, including Hessian matrix building, scale space generation, primary determination of feature points, precise positioning of the feature points, main direction determination of selected feature points, feature point descriptor construction and feature value string generation; document image feature value comparison, including document similarity calculation and comparison algorithm optimization. According to the method, an image is preprocessed through software, and additional hardware equipment does not need to be added. According to the method, the scale-invariant feature is creatively introduced for improving a typical SURF feature extraction algorithm, so that the problem of matching failure due to error amplification of the SURF algorithm caused by scale variations is fundamentally solved. The method has the advantage that a multi-thread technology and a large cache are used for solving the problems of large data volume calculation during comparison and the harsh time requirement of a user on an electronic government affair platform.

Description

technical field [0001] The invention belongs to the technical field of computer image recognition, and in particular relates to a fast document type recognition method based on full-width feature extraction. Background technique [0002] The e-government platform is the frontier of the government's external informatization office. Each government department needs to receive a large number of forms and copy materials submitted by users every day. If these materials are manually identified by document type, it will take a lot of manpower and delay processing time. , it is difficult to effectively classify and manage the materials submitted by users, and more advanced identification technology is undoubtedly in urgent need. [0003] At present, the document recognition software running on the e-government platform mainly uses OCR to assist in some text recognition work, so as to reduce the text entry work of the staff. However, how to quickly identify whether the materials wit...

AI Technical Summary

Problems solved by technology

[0006] The present invention aims at the problem that the existing document type recognition method becomes poor or even completely invalid when the document image to be recognized has noise, uneven illumination, image rotation and curl deformation, etc., and proposes a method based on full-frame feature extraction. A fast document type recognition method, which can effectively solve the impact of illumination, noise, deformation, etc. on document type recognition, and has good robustness to rotation, curling and other phenomena, and is accurate in various lighting environments. At the same time, the calculation speed of this method is fast, which can meet the occasions with high real-time requirements

Images