A SMART LIBRARY INVENTORY SYSTEM BASED ON BOOK PINE TEXT RECOGNITION

Abstract

Efficient management of library resources is essential for educational institutions, research centers, and public libraries. Traditional library inventory systems rely heavily on manual verification or barcode scanning techniques, which require significant human effort and time. Although technologies such as RFID have improved automation, their implementation cost remains high for many institutions. This research proposes a smart library inventory system that uses computer vision and optical character recognition (OCR) techniques to identify books directly from the text printed on their spines. The system captures images of bookshelves using a camera, detects book spines from the image, and extracts textual information such as book titles and author names. The extracted text is the matched with a digital library database to verify the presence and location of books. This automated approach minimizes manual effort, improves inventory accuracy, and speeds up the library management process. The proposed system demonstrates how artificial intelligence and image processing can be integrated into library management systems to enhance efficiency and reliability. The study also discusses system design, implementation methods, challenges, and potential improvements for future research.