Enhancement and Tampering Detection for Surveillance System

Surveillance systems are prevalent in our lives, and surveillance videos are often used as significant evidence for judicial forensics. Good quality surveillance videos have become essential, due to the increase in the rate of malpractice and misconduct. therefore, there arises a need for high-resolution videos in law enforcement. Along with bad quality, there is one more problem that is faced that is of low light in the image or the video. To tackle these problems, an integrated system ResMaster is proposed which would be a web application, wherein along with video enhancement, video tampering will also be detected. Video enhancement will be done using exposure fusion framework which works on the pixels depending on their exposure. With the use of this framework, results are obtained with enhanced regions that were previously overexposed or underexposed, whereas tampering will be detected using correlation of noise residue for which Gaussian Mixture Model will be used.