Building book catalog using image processing

Abstract

In this article we describe an automatic initialization-free method for building and managing book catalogs. At the heart of the system there is an image processing algorithm which can analyze the contents of an image taken of a bookshelf, separate the book spines into subsequent images and use an OCR algorithm on it to find the title and author of the book. The title and author information is searched for in public book databases to find all the details of the given book which can then be stored. The article describes the methods used along with accuracy measurements for the different modules of our system.