Detection and identification method oriented to image tablet shot by mobile terminal

Abstract

The invention provides a detection and identification method oriented to an image tablet shot by a mobile terminal. The method mainly comprises the following steps: separating a tablet area from a hybrid document image shot by the mobile terminal to obtain a tablet image by virtue of a connected component analysis method, and performing image correction operation on the tablet image; detecting and positioning a straight line in the corrected tablet image by virtue of rough to detailed straight line detection algorithms, performing correction processing on the detected straight line, and extracting information of each tablet cell in the tablet image; performing text identification on the extracted information of the tablet cells, and reconstructing a corresponding tablet in a document by virtue of a cell-combination-based tablet creation algorithm. The method is highly adaptive to regular and irregular tablet creation, and conversion operation from the image tablet to a document file can be finished after simply editing and modifying a created word document, so that the office automation efficiency is improved, and high practicability is achieved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image table detection and recognition method for shooting by a mobile terminal. Background technique [0002] In recent years, the rapid progress of science and information technology, especially the continuous development of computer technology, network technology and mass storage technology, makes us come into contact with a large amount of information every day, so we in the information age usually need to use computers to process There is a large amount of digital document information and document images, but this requires us to complete the transformation process from paper information to digital information to facilitate subsequent computer image processing and document recognition. [0003] The conversion process of paper documents to electronic documents mainly includes: using mobile terminals or scanners to complete the collection of document images, identify...

AI Technical Summary

Problems solved by technology

However, when using a mobile terminal device to shoot an image table, due to the shooting angle, the collected image is prone to image distortion, which will have a great impact on the text recognition step. At the same time, the image distortion will also cause The table lines in the captured image will be seriously deformed. Therefore, the above-mentioned unrestricted table detection and recognition algorithm has considerable challenges in the detection and recognition of the image table captured by the mobile terminal. From the current recognition effect From the above point of view, the final recognition effect is hardly satisfactory.

Images