Deep Convolutional Neural Network based Image Steganogrpahy Technique for Audio-Image Hiding Algorithm
D.Rathna Kishore1, D.Suneetha2, P.Narendra Babu3, P.Chinababu4
1Dr.D.Rathna Kishore*, Professor, Department of CSE, NRIIT, Vijayawada, A.P, India.
2Dr.D.Suneetha, Professor, Department of CSE, NRIIT, Vijayawada, A.P, India.
3Mr.P.Narendra Babu, Associate Professor, Department of CSE, NRIIT, Vijayawada, A.P, India.
4P.Chinababu, Assistant Professor, Department of CSE, LIMAT, Vijayawada, A.P, India.
Manuscript received on March 29, 2020. | Revised Manuscript received on April 27, 2020. | Manuscript published on April 30, 2020. | PP: 2187-2189 | Volume-9 Issue-4, April 2020. | Retrieval Number: D7843049420/2020©BEIESP | DOI: 10.35940/ijeat.D7843.049420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Steganography is one expanding filed in the area of Data Security. Steganography has attractive number of application from a vast number of researchers. The most existing technique in steganogarphy is Least Significant Bit (LSB) encoding. Now a day there has been so many new approaches employing with different techniques like deep learning. Those techniques are used to address the problems of steganography. Now a day’s many of the exisiting algorithms are based on the image to data, image to image steganography. In this paper we hide secret audio into the digital image with the help of deep learning techniques. We use a joint deep neural network concept it consist of two sub models. The first model is responsible for hiding digital audio into a digital image. The second model is responsible for returning a digital audio from the stego image. Various vast experiments are conducted with a set of 24K images and also for various sizes of images. From the experiments it can be seen proposed method is performing more effective than the existing methods. The proposed method also concentrates the integrity of the digital image and audio files.
Keywords: Steganography, Least Significant Bit, Deep Learning, Deep Convolutional Neural Network.