Digital Image Forgery Detection using Wavelet Decomposition and Edge Detection (original) (raw)
In the presented work, a copy-paste or cut-paste detection algorithm is discussed. The algorithm is based on wavelet decomposition of the input image and then extracting the edges so that the cut-paste or copy-paste region is detected by examining the edge pixels in wavelet domain. The scheme consists of: image acquisition, gray scale conversion, wavelet decomposition using haar wavelet, edge detection using sobel filter and then analysis of the region falling under the straight edge chain. The cut-paste or copy-paste region is normally in rectangular or square edge region. The edge pixels are analyse for their chain code so that the straight line could be extracted from the pixel tracing. This gives the possible region of image forgery. Further, the input image and forged images are compared with respect to different properties like entropy, power energy, standard deviation etc and re discussed in the algorithm.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.