Applying General Regression Neural Network in Digital Image Watermarking

Combing the human visual system with the image local correlation properties, a robust digital image watermarking method based on general regression neural network (GRNN) and fuzzy c-mean clustering algorithm (FCM) is proposed. It uses FCM to adaptively identify watermarking embedding location and strength and uses trained GRNN to help watermarking embedding and extracting. In watermark extracting it does not need original image. Experimental results show that the proposed method has better performance than the similar method in countering common image process, such as JPEG compression, noise adding, filtering and so on.