Method for classified search of house property file image

Abstract

The invention discloses a method for classified search of a house property file image, which is characterized by comprising the following steps of: (1) preprocessing of the house property file image, including normalization, binarization, denoising and edge cutting, thereby obtaining a preprocessed image; (2) expansion, inversion and refining of the preprocessed image, extraction of paragraph features, and formation of comprehensive features in combination with local pixel features to commonly represent image features; (3) classified training of the comprehensive features with an SVM (support vector machine) to obtain a classification model, wherein an RBF (radial basis function) kernel function is adopted in training, and an SMO (sequential minimal optimization) algorithm is adopted as a classification algorithm; and (4) classified search of the file image through the classification model. The method has the beneficial effects of high accuracy of classification and strong pertinence.

Description

technical field [0001] The invention relates to an electronic and digital management technology of paper documents, in particular to a method for classifying and retrieving real estate document images. Background technique [0002] Real estate document images refer to a special type of images with text and charts as the main content, which are mainly document images converted from real estate paper files through scanners and other equipment. General real estate documents only have a few fixed categories: contracts, invoices, copies of ID cards, etc. Each category has the same format. This kind of image data is different from general natural scene images in terms of content, feature extraction and retrieval targets. Therefore, the real estate document image classification is different from other image classifications in that the real estate document image classification does not need to really identify the character content of the two documents, but can judge whether the tw...

AI Technical Summary

Problems solved by technology

Due to the particularity of the real estate file itself, the above methods cannot represent the paragraphs of the real estate file well, and the extracted features cannot reflect the document itself well, resulting in inaccurate classification retrieval results

Moreover, when scanning, due to different scanned documents, the size of the image may be different, and the image size of the same document after different scans will also be different, which will also affect the accuracy of classification retrieval

Images