Abstract：An Autonomous Vision-Based Sheelf-Reader Robot USING FASTER R-CNN Purpose The purpose of this study is to designprogram and implement an intelligent robot for shelf-readingan essential task in library maintenance is shelf-readingwhich refers to the process of checking the disciplines of books based on their call numbers to ensure that they are correctly shelvedShelf-reading is a routine yet challenging task for librariansas it involves controlling call numbers on the scale of thousands of books promptlyDesignmethodologyapproach Leveraging the strength of autonomous robots in handling repetitive tasksthis paper introduces a novel vision-based shelf-reader robotcalledemphParsand demonstrate its effectiveness in accomplishing shelf-reading tasksAlsothis paper proposes a novel supervised approach to power the vision system ofemphParsallowing it to handle motion blur on images captured while it movesAn approach based on Faster R-CNN is also incorporated into the vision systemallowing the robot to efficiently detect the region of interest for retrieving a books informationFindings This paper evaluated the robots performance in a library with120000 books and discovered problems such as missing and misplaced booksBesidesthis paper introduces a new challenging data set of blurred barcodes free publicly available for similar research studiesOriginalityvalue The robot is equipped with six parallel cameraswhich enable it to check books and decide moving pathsThrough its vision-based systemit is also capable of routing and tracking paths between bookcases in a library and it can also turn around bendsMoreoveremphParsaddresses the blurred barcodeswhich may appear because of its motionLibrary AutomationThis robot is designed to read on the automation frameThis is an important part of ensuring the correct placing of booksThis researchprogrammingand implementing a smart robot for reading on the framethis is a basic task in library maintenanceThe study mentioned that using the advantages of the autonomous robot in handling duplicate tasksthis article introduces a new type of visual-based reading robot nameEMPHParsEffectIn additionthis article also proposes a new monitoring method to driveEMPHParsvisual systemso that it can process the motion blur on the image captured when movingThe method based on Faster R-CNN is also included in the visual systemallowing robots to effectively detect interesting areas to obtain books to obtain books