Automatic classification method of old books

Abstract

The present invention relates to an automatic classification method of old books. The method comprises the following steps of: the step 1: performing training of a book classifier; the step 2: performing training of a book model, extracting book feature values, and putting data into a feature database; the step 3: detecting books by employing the classifier; the step 4: obtaining image data of target books; the step 5: inputting the image data into the model to obtain book data; and the step 6: classifying different books. The book classification method is an effective integration of machine learning and vision algorithm, can obtain detailed information of the books in the process of book image collection, and can performs classification. The method breaks through the limitation of manualsorting, is convenient to operate, can effectively save cost and manpower and materials, is high in transplantability, can integrate a plurality of intelligent products, and can create a certain economic benefit and practical value.

Description

Technical field: [0001] The invention relates to the field of book classification, in particular to an automatic classification method for old books. Background technique: [0002] At present, there is a large market for old books, and the most common way to classify old books is to use manual classification, which is time-consuming and labor-intensive. Especially for some large libraries, the classification of books occupies a large amount of working time of librarians, resulting in low work efficiency of managers. Especially for the classification of second-hand books, its classification is a huge workload, and it also hinders the recycling of second-hand books. [0003] The artificial classification method has great limitations, and for some specific occasions, the efficiency is low and the operation is not convenient enough. Contents of the invention [0004] The prior art cannot satisfy people's needs, and in order to make up for the deficiencies of the prior art, t...

AI Technical Summary

Problems solved by technology

[0002] At present, there is a large market for old books. The most common way to classify old books is to use manual classification. This classification method is time-consuming and labor-intensive.

Especially for some large libraries, the classification of books takes up a lot of working time of librarians, resulting in low work efficiency of managers

Especially for the classification of second-hand books, the classification is a huge workload, which also hinders the recycling of second-hand books

[0003] The manual classification method has great limitations. For some specific occasions, the efficiency is low and the operation is not convenient enough.

Images