Novel Fragile Watermarking Scheme using an Artificial Neural Network for Image Authentication
Yu‐Cheng Fan, Yu-Yao Hsu · 2015
This paper presents a novel fragile watermarking scheme based on an artificial neural network (ANN). The fragile waterma rk is designed according to the characteristics of the origina l image. If the image is modified, the alteration can be detect ed via the fragile watermark without original image. Based on the type of alteration, we can determine what modifications have been perform ed. An artificial neural network is used to analyze the modification s. The experimental results show that the proposed method can detect tampering, locate where the tampering has occurred, and recognize what kind of alteration has occurred. This method has a high success ratio in recognizing the types of modifications and p rovides sufficient evidence. The experimental results demo nstrate that our method is indeed effective.