Form image classification method

Abstract

The present invention relates to a form image classification method. For training images, first, forms, belonging to the same type, are averaged to obtain a mean value image, each pixel point of the mean value image is a mean value of the pixel point in each training image, and each type of a mean value template is composed of the obtained mean value images; then three weight values including a consistency weight, a randomness weight and a vibration weight are solved; when form classification is carried out, the three weights and the mean value template are used for carrying out classification calculation. The average forms can be changed to the forms of a position pixel point mode of the position; then a variance and different weight value are calculated aiming at the forms of the mode. By virtue of the form image classification method, the influence caused by the randomness of user filling-in information can be reduced, and the importance of distinguishing information in form layout can be amplified simultaneously, so that a very good classification performance is obtained aiming at the Chinese forms with similar layout.

Description

technical field [0001] The invention belongs to the technical field of document classification and pattern recognition, and in particular relates to a form image classification method based on distance measurement. Background technique [0002] At present, in many businesses (such as banking, insurance, statistics, etc.), a large number of Chinese forms are generated by printing / copying, etc., and then passed to customers for printing or manual filling, resulting in a large number of Chinese forms existing in paper form , which brings many challenges and difficulties to the later form automation processing. On the other hand, in order to make the office more automated, and then be able to extract and mine useful information from the form, the demand for form automation is increasingly strong. [0003] The automated processing of forms usually includes a series of processes such as scanning, reading, classification, layout analysis, identification and editing of paper forms....

AI Technical Summary

Problems solved by technology

[0017] The existing form classification technology mainly solves the classification problem of forms with different layouts in the form classification problem, but for forms with similar or similar layouts, this type of algorithm considers that they belong to the same type of form

Therefore, in order to solve the problem that similar forms are difficult to classify, the present invention proposes a simple and effective Chinese form classification method based on weighted distance, which reduces the impact of the randomness of the information filled in by users, and at the same time enlarges the importance of distinguishing information in the form layout. In order to achieve better classification performance for Chinese forms with similar layouts

Images