Spiking neural network based scrambled watermark hiding in low-frequency region of digital image
Sunesh Malik, R. Rama Kishore · Journal of Information and Optimization Sciences · 2020
The digital image watermarking system has proven its own efficiency without any doubt to establish authenticity and to prevent the misuse of digital images in today’s world. In this process, the present paper proposes a new and significant Spiking Neural Networks based image watermarking method (SNNW) with the objectives of robustness and security alongwith less time complexity. In this proposed scheme, the extraction procedure is taken as an optimization problem which is solved by implementing the spiking neural networks SNN. The proposed SNNW employs spiking neural networks with the aim of achieving more robustness against the different attacks along with less time complexity. In addition to this, security of proposed SNNW method is obtained by exploiting chaotic based scrambling on watermark. As a result, a series of experiments are conducted and carefully analyzed on a set of images and assessed in the form of peak signal to noise ratio PSNR, Normalization correlation NC, extraction rate and time complexity. The experimental results of SNNW system are carefully analyzed and found robust and secure to various attacks like histogram equalization, compression, noising and filtering attacks. Moreover, effectiveness of proposed SNNW method is exhibited through a comparison with different state of the art methods.