| In recent years, our country has more kinds of examinations, and we have seen a rapid increase in the frequency of examinations. And the same time, the number of examiners is also rapidly increasing year by year. There are exams for Occupational qualifications, education degree, foreign language, civil servants, computer etc.With the development of science and technology, video surveillance system has been widely used in various types of test, thereby generating a massive examination surveillance video. In line with the principle of fairness and justice, we need to review the examination surveillance video after the exam. However, the review of video is a huge workload, which brings many problems, mainly include the management issues and the review issues.Based on the above problems, this paper designs and implements the exam surveillance video processing system based on Hadoop platform, and the exam surveillance video processing system including two child module, video management module and video processing module. The management issues is solved by video management module. By using the Hadoop platform,we stored the examination surveillance video in Hadoop Distributed File System. The main functions of the modules are:personnel information management,the video upload, the video download and video processing module link.The review issues is solved by video processing module. The module can automatically capture and storage in the suspected cheating in the examination surveillance video, make it convenient way for auditors to review the video. The main function of this module is automatically capture the suspected cheating, in order to achieve this function, we propose and implement an algorithm based on OpenCV which is used to automatically screen video clips which may contain cheating in the examination.The intelligent system can reduce the burden of the censors, At the same time, it also can improve the efficiency of video review. |
PDF Full Text Request