A New Image Watermarking Technique based on Random Forests
Sanping Li · 2018
Random Forests" is an algorithm developed by Breiman and Cutler in 2001 [1] .It runs by constructing multiple decidion trees while training and outputting the classes that is the mode of the classes output by individual trees.It has improved performance over single decision trees, and it is much more efficient than traditional machine learning techniques, e.g.artificial neural networks and support vector machine.In this paper, a new image watermarking technique based on Random Forests (RF)is proposed.First, the image is decomposed through discrete wavelet transform.Then we use the relationship between the selected coefficient and its neighboring coefficients to train RF.Thanks to the good learning ability of RF, the watermark is adaptively embedded in wavelet domain and also can be extracted by the well trained RF.Experimental results show that the algorithm is robust against high intensity JPEG, JPEG2000, geometric distortion.