Abstract：Machine Assistance in Collection BuildingNew ToolsResearchIssuesand Reflections PurposeBook sorting system is one of specific application in smart library scenariosand it now has been widely used in most libraries based on RFIDradio-frequency identification devicestechnologyBook identification processing is one of the core parts of a book sorting systemand the efficiency and accuracy of book identification are extremely critical to all librariesIn this paperthe authors propose a new image recognition method to identify books in libraries based on barcode decoding together with deep learning optical character recognitionOCRand describe its application in library book identification processingDesignmethodologyapproachThe identification process relies on recognition of the images or videos of the book cover moving on a conveyor beltBarcode is printed on or attached to the surface of each bookDeep learning OCR program is applied to improve the accuracy of recognitionespecially when the barcode is blurred or fadedThe approach the authors proposed is robust with high accuracy and good performanceeven though input pictures are not in high resolution and the book covers are not always verticalFindingsThe proposed method with deep learning OCR achieves best accuracy in different verticalskewed and blurred image conditionsResearch limitationsimplicationsMethods that the authors proposed need to cooperate and practice in different book sorting machineSocial implicationsThe authors collected more than 500 books from a libraryThese photos display the cover of more than 100 randomly picked books with backgrounds in different colorseach of which has about five different pictures captured from variety anglesThe proposed method combines traditional barcode identification algorithm with the authorsmodification to locate and deskew the imageAnd deep learning OCR is involved to enhance the accuracy when the barcode is blurred or partly fadedBook sorting system design based on this method will also be introducedOriginalityvalueExperiment demonstrates that the accuracy of the proposed method is high in real-time test and achieves good accuracy even when the barcode is blurredDeep learning is very effective in analyzing image contentand a corresponding series of methods have been formed in video content understandingwhich can be a greater advantage and play a role in the application scene of intelligent libraryLibrary AutomationAutomatic recognition and barcodes of books through deep learning technologyThis study puts forward a new barcode-based and deep-learning optical character recognitionOCRimage recognition method to identify books in the libraryWhen the barcode is blurred or fadedthe deep learning OCR program is applied to improve the accuracy of recognition