Abstract：Research on classification and identification of library based on artificial intelligence With the popularization of open borrowingit is urgent for major libraries to explore new collection classification and identification systems to assist or replace librarians in their daily book cataloguing workThe messy books can be marked according to invisible colorswhich puts forward a new method of library classificationIn the algorithm used to detect image edgesby looking for the traditional sensitivity measurement systemthe best parameters are applied to the wrong book image edgesand the test results are verified to find the best parameter rangeA two-dimensional algorithm is used to segment the information element matrix to reduce soundThe results show that the image edge test is effective and reduces the influence of noise on the image testThe self-adjustment control algorithm of a single neuron is used to control the robot orbitwhich makes the robot not only react quicklybut also reduce a small part of the error in meeting the actual needs of the robotThe effect on the result is getting smaller and smaller Collection ClassificationThe article mentioned the use of new collection classification and identification systems to help or alternate libraries to perform daily booksThis is based on the new method mentioned-the mark of the book based on the invisible colorRobotThe article mentioned that the self-adjustment control algorithm using a single neuron to control the rails of the robot enables the robot to respond quickly and reduce a small part of the error in meeting the actual needs of the robotMode recognitionHere is the use of algorithms to detect the edge of the image and reduce the impact of noise