Design And Implementation Of Content-based Confidential Information Monitoring System

Government agencies,large organizations,enterprises and other units have a lot of classified information that needs strict management.With the fast progress of informatization,a large number of classified information is electronic,which makes it extremely difficult to manage classified information,increasing the risk of leaks significantly.Firstly,this thesis introduces the design objectives and technical difficulties of the monitoring system:Quick word matching and monitoring high-precision OCR recognition are the key technology,otherwise,the stability of the scanning core is the guarantee of the systemThen,the thesis describes requirements analysis modeling and system design and proposes to integrate dual process document scanning method that the task of classified information recognition and interface are each packaged as a process in which Quick word matching and monitoring high-precision OCR recognition are used.In addition,the thesis also optimizes the following algorithms:1.SQLite-based online symmetrical high-speed hot standby database is proposed to improve the reliability of the system core data;2.A fast keyword matching algorithm for large number of long key words is proposed by using Trie structure;3.A local adaptive binarization algorithm is proposed to improve the efficiency of the classified image binarization;4.An image tilt correction algorithm of high robustness and high precision is proposed to improve the detection precision of the classified information in tilted images;5.Low resolution OCR algorithm is proposed to promote the recognition accuracy of classified information in low resolution image;It is obtained by quantitative experiments that the stability of the system is superior to single-process and the image tilt correction algorithm,the local adaptive binarization algorithm and low resolution OCR algorithm have a clear advantage over other existing algorithms in terms of speed and accuracy,etc.Finally,this paper completes the development of the monitoring software system,realizes the monitoring in copying,communication and transformation of classified information in the computer according to the anti-virus software principal,and completes document scanning,USB device monitoring,Internet access monitoring and other functions.A content-based classified information monitoring system with high stability based on the related technology are achieved.This monitoring system can manage the classified information for government and enterprises.It can also monitor dissemination and diffusion of classified information automatically,accurately and timely to prevent leakage incidents,filling the gaps in relevant applications.