我原来做对比实验的6篇文献分别来自Nonlinear Dynamics，Neurocomputing， Multimedia Tools & Applications， Optics Communications。而审稿人3叫我做对比实验的文章分别是下面7篇（怎么看这7篇文章是一伙的）：
 Khan J S, Ahmad J, Khan M A. TD-ERCS map-based confusion and diffusion of autocorrelated data[J]. Nonlinear Dynamics, 2016, 87(1):1-15.
 Kanso A, Ghebleh M. An efficient ( t, n )–threshold secret image sharing scheme[J]. Multimedia Tools & Applications, 2016, 76(15):1-20.
 Ahmad J, Khan M A, Hwang S O, et al. A compression sensing and noise-tolerant image encryption scheme based on chaotic maps and orthogonal matrices[J]. Neural Computing & Applications, 2016:1-15.
 Kanso A, Ghebleh M. An algorithm for encryption of secret images into meaningful images[J]. Optics & Lasers in Engineering, 2017, 90:196-208.
 Rehman A U, Khan J S, Ahmad J, et al. A New Image Encryption Scheme Based on Dynamic S-Boxes and Chaotic Maps[J]. 3d Research, 2016, 7(1):1-8.
 Ahmad J, Khan M A, Ahmed F, et al. A novel image encryption scheme based on orthogonal matrix, skew tent map, and XOR operation[J]. Neural Computing & Applications, 2017(3):1-11.
 Khan J S, Ahmad J. Chaos based efficient selective image encryption[J]. Multidimensional Systems & Signal Processing, 2018(4):1-19.
Reviewer #3: The authors proposed "文章的题目". It is interesting. However, I have some questions as follows:
1) There are many typing and grammatical errors in several places listed as follows:
2) I wish there are comparisons with others. The other representative methods such as: based on TD-ERCS based confusion and diffusion, Dynamic S-Box, DNA Compression Sensing, etc. The papers are related with these methods please refer to Nonlinear Dynamic (2017) 87: 93. doi.org/10.1007/s11071-016-3028-2; Multimedia tools and application 10.1007/s11042-016-3917-x; Neural Computing and Applications, doi.org/10.1007/s00521-016-2405-6; Optics and Laser in Engineering 10.1016/j.optlaseng.2016.10.009; 3D Research, doi.org/10.1007/s13319-016-0084-9.
3) The scheme is a bit complex, therefore, add flowchart for your proposed scheme.
4) Irregular deviation and Maximum deviation should be done and compared with the following articles. Kindly refer to Neural Computing and Applications, doi.org/10.1007/s00521-017-2970-3.
5) There are many related studies, such as Multidimensional Systems and Signal Processing, https://doi.org/10.1007/s11045-018-0589-x; on the encryption algorithm based on the chaotic system in recent years, in order to show your innovations, contrast tests should be done and compared with these articles.
At last, I suggest the authors must carefully handle the above comments. If having good revision, then I am willing to review again.