Research And Implementation Of Automatic Equipment Information Collection In Intelligent Machine Room Management System

With the speedy advancement of Internet technology,the required computing performance and storage capacity have increased significantly,resulting in the continuous expansion of computer room scale and accelerated equipment replacement,making it more difficult to manage computer room assets.One of the important tasks in the asset management of computer room is the collection of equipment information.The usual way for computer room managers is to check the equipment labels and manually enter the equipment information on the labels into the system,which is ineffective and susceptible to errors.The equipment information can be divided into the information of the equipment itself and the location information of the equipment on the rack.There is device information in the device image,which can be obtained by analyzing the image.There may be a variety of devices in a device image,so it is necessary to obtain the information of all devices.Compared with the whole image,the proportion of scale indication is very small,so it is difficult to locate it.In the actual shooting process,the deviation of shooting angle will lead to the deformation of equipment image,and environmental factors will also affect the image quality.This thesis deeply studies the characteristics of the device image and the technology of machine vision.Relying on the method of deep learning,it puts forward the detection algorithm of device attribute information and device location information,and designs and realizes the automatic acquisition subsystem of device information in the intelligent computer room management system.Through the device information automatic collection subsystem,users only need to take a picture of the device and upload it,they can get the information of all the devices in the picture.After the user reviews the test results,upload them to the intelligent computer room management system to complete the collection of equipment information.First of all,this thesis introduces the research background and related technologies involved in the equipment information automatic acquisition subsystem,and proposes the equipment attribute information detection algorithm and equipment location information detection algorithm.Secondly,the requirements of the subsystem are analyzed,the functional modules are divided,and the overall architecture is designed.After that,the subsystem is designed and implemented in detail,and the function,performance and accuracy of the subsystem are tested by "black box method".Finally,the full thesis is summarized,describing the shortcomings of the study and pointing out the directions for improvement.