Exposing Double MPEG Compression Based on First Digit Features
Wang Wan① Jiang Xing-hao①② Sun Tan-feng①②
①(School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China) ②(National Engineering Laboratory for Information Content Analysis Technologies, Shanghai 200240, China)
Abstract：Video forgeries can be effectively detected by discriminating double MPEG compression, in order to authenticate the truism and reliability of digital videos. Through delving into the fitting results of the first digit distribution of Alternating Current (AC) coefficients with logarithmic law, a double MPEG compression detection algorithm is proposed by extracting a 12-D first digit feature and employing Support Vector Machine (SVM). Experiments show that this algorithm has high accuracy and efficiency.
2012, Vol. 34 
