The Robust Digital Image Watermarking Scheme with Back Propagation Neural Network in DWT Domain

This paper presents a blind robust digital image watermarking approach based on back propagation neural network in DWT domain. The cover image is decomposed up to 4-levels using DWT. The bitmap of size 64x64 is selected as a watermark. The Back Propagation Neural Network (BPNN) is implemented while embedding and extracting the watermark. In BPNN the errors will be back propagated to the input layer, so that the weights and learning rate parameters may be changed to get the output. The BPNN is trained in such a way that it converges fast and reaches high accuracy. The proposed watermarking algorithm is imperceptible and robust to some normal attacks such as JPEG compression, salt and pepper noise, rotation, median filtering and cropping.