Font Size: a A A

Design And Implementation Of Automatic Audit Of Teaching Resources In The Teaching Cloud Platform

Posted on:2022-06-06Degree:MasterType:Thesis
Country:ChinaCandidate:X Y ChenFull Text:PDF
GTID:2507306338985649Subject:Computer technology
Abstract/Summary:PDF Full Text Request
In recent years,with the spread of the global outbreak and the development of information technology,the online teaching mode has thrived,making the online teaching platform has been effectively promoted and applied.As a result,the security of teaching resources becomes more and more important.The teaching platform needs to check the content of the teaching resources uploaded by users to ensure that the teaching resources comply with relevant national laws and regulations.At present,the teaching cloud platform mainly relies on manual work to complete the content check of teaching resources on the platform.On account of high cost,low efficiency and no unified audit standard of manual check,the teaching cloud platform urgently needs an automatic audit system with excellent performance and high efficiency,which is able to check various forms of teaching resources accurately.Therefore,aiming at teaching resources of video,image and text types,this thesis designs and implements an automatic audit system of teaching resources,to solve the key point of manual audit of teaching resources on the teaching cloud platform.The research work of this thesis mainly includes the following two aspects:(1)Investigating the related technologies and the existing audit scheme.Combined with the business requirements of the teaching cloud platform,the automatic audit system for teaching resources includes various modules,which contains check of sensitive words for text resources,check of pornographic contents for image resources and check of violence for video resources.And Hash algorithm is used to complete the fingerprint maps database,for decreasing the burden and increasing the efficiency of audit.(2)Investigating the existing audit algorithm for violence based on video.Comparing the merits and demerits of traditional recognition model combined with feature extraction and classifier and deep learning model,extracting the time-space relation by Histogram of Oriented Gradient(HOG)and Motion Scale Invariant Feature Transform(MoSIFT)and Color Layout Descriptor(CLD),this thesis designs and completes the violence audit module of video resources.Finally,this thesis tests the automatic audit system of teaching resources roundly,and the test results effectively verify the effectiveness and feasibility of the system.The method of automatic audit put forward by this thesis has a certain reference value for the security audit of contents in teaching resources.
Keywords/Search Tags:content security, automatic audit, violence audit, pornographic image recognition
PDF Full Text Request
Related items