Most existing image blurring forgery detection algorithm consider only one single feature introduced by blurring operation

in order to improve algorithm detection efficiency

propose blur forgery detection algorithm based on kernel principal component analysis（ KPCA）. Through singular value decomposition（ SVD）

extract the first group of features

calculate the secondlary fuzzy correlation as the second group of features

calculate image quality factor as the third group of feature. Multi-feature fusion are achieved using KPCA. Judgement is carried out using support vector machine

so as to realize blur forgery detection. Experimental results show the proposed algorithm can effectively detect blur operation trace of digital tampering image and can accurately locate blur forgery area.