IEEE Consumer Electronics Magazine - September 2018 - 45

Table 1. A comparison of secret images with
shared images.
Secret and
Shared Image

Correlation

MSE

PSNR (dB)

SI, S 1

0.0003

14,455.15

6.56

SI, S 2

-0.0062

10,232.37

8.07

SI, S 3

0.0036

9,616.29

8.33

SI, S 4

0.0035

9,638.36

8.32

SI, S 5

0.0001

9,799.87

8.25

Table 2. A comparison of a secret image with stego
in combination with recovered cover images.
Secret, Stego, and
Recovered Cover Image

Correlation

MSE

PSNR (dB)

SI, ST

0.0130

9,588.17

8.35

SI, (ST, R 1)

-0.0289

10,050.06

8.14

SI, (ST, R 1, R 2)

0.0282

9,546.88

8.14

SI, (ST, R 1, R 2, R 3)

-0.0588

10,529.23

7.94

SI, (ST, R 1, R 2, R 3, R 4)

0.9998

1.99

45.77

only secure but also helped a user trap and mislead an attacker. The concept of secret sharing was used to provide better
security and to deceive the attacker by giving pseudo images
as a result. Steganography was used to conceal secret images
into a stego image.
The limitations of the proposed scheme are 1) the proposed scheme uses a division operator that reduces pixel size,
allowing a user to use this algorithm for fewer than 255 images
at one time (if more than 255 are used, the algorithm will not
work) and 2) the stego image comes out to be random, which
could alert an intruder that they are being deceived. In future
tests, we will use an alternative for the division operator. We
will make a stego image systematic so that an intruder will
not know they are being tricked.

ABOUT THE AUTHORS
Mohit Rajput (mrajput7@gatech.edu) is with the Georgia
Institute of Technology, Atlanta.
Maroti Deshmukh (marotideshmukh@nituk.ac.in) is with
the National Institute of Technology, Uttarakhand, India.
Neeta Nain (nnain.cse@mnit.ac.in) is with the Malaviya
National Institute of Technology Jaipur, India.
Mushtaq Ahmed (mahmed.cse@mnit.ac.in) is an assistant professor in the Department of Computer Science
and Engineering, Malaviya National Institute of Technology Jaipur, India.

REFERENCES
the secret image. The MSE and PSNR for the pair came out
to be . 1.99 and 45.17 dB, respectively, which indicates both
images were nearly identical. But all of the other combinations of the stego image and the recovered cover images with
secret images showed the exact opposite result, as the correlation value came out to . 0, with the root MSE (RMSE) and
PSNR being greater than 9,500 and fewer than 8.35 dB,
respectively. This indicates that the other combinations were
completely different from the secret image. The < n shared
images in combination with the stego image did not reveal
any information about the secret image. The results obtained
from applying quality measures on the secret image, the
stego image, and the recovered cover images are illustrated in Table 2.
The time complexity increased with the increase of the
cover images. The time complexity T was dependent on
image dimension (D), image type (I ), and the number of
cover images (N ), T ? (D # I # N). The space complexity
was dependent on the cover images, i.e., as the cover images
increased, the space complexity increased. Due to the two
levels of encryption, the space complexity was high. To
reduce the space complexity and increase the security of
shared and recovered cover images, we will disseminate the
shared and recovered cover images into different servers.

CONCLUSION
In this article, we proposed a novel encryption technique
using secret sharing and steganography. This scheme was not

[1] S. P. Mohanty, A. Sengupta, P. Guturu, and E. Kougianos, "Everything you want to know about watermarking," IEEE Consum. Electron.
Mag., vol. 6, no. 3, pp. 83-91, July 2017.
[2] K. Nimmy and M. Sethumadhavan, "Novel mutual authentication
protocol for cloud computing using secret sharing and steganography," in
Proc. IEEE Fifth Int. Conf. Applications Digital Information Web Technologies, 2014, pp. 101-106.
[3] M. Naor and A. Shamir, "Visual cryptography," in Proc. Workshop
Theory Application Cryptographic Techniques (EUROCRYPT), Perugia,
Italy, 1994, pp. 1-12.
[4] M.-J. Li and J. S.-T. Juan, "A flexible multiple-secret image sharing
scheme by shifting random grids," in Proc. IEEE 17th Int. Symp. Consumer Electronics (ISCE), 2013, pp. 289-290.
[5] M. Rajput and M. Deshmukh, "Secure (n, n+ 1)-multi-secret image
sharing scheme using additive modulo," Procedia Comput. Sci., vol. 89,
pp. 677-683, Jan. 2016.
[6] Y. Tian, J. Ma, C. Peng, and Q. Jiang, "Fair (t, n) threshold secret
sharing scheme," IET Information Security, vol. 7.2, pp. 106-112, June
2013.
[7] T.-H. Chen and C.-S. Wu, "Efficient multi-secret image sharing
based on Boolean operations," Signal Processing, vol. 91.1, pp. 90-97,
Jan. 2011.
[8] L. J. Anbarasi and S. Kannan, "Secured secret color image sharing
with steganography," in Proc. IEEE Int. Conf. Recent Trends Information
Technology (ICRTIT), 2012, pp. 44-48.
[9] C.-C. Chang, Y.-H. Chen, and L.-Y. Chuang, "Meaningful shadows
for image secret sharing with steganography and authentication techniques," J. Inform. Hiding Multimedia Signal Processing, vol. 5, no. 3,
pp. 342-352, 2014.
[10] M. R. Ogiela and K. Koptyra, "False and multi-secret steganography in digital images," Soft Computing, vol. 19.11, pp. 3331-3339,
Nov. 2015.

september 2018

IEEE Consumer Electronics Magazine

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September 2018