See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325554858

A fusion steganographic algorithm based on faster R-CNN

Article in Computers, Materials & Continua · January 2018

DOI: 10.3970/cmc.2018.055.001

CITATIONS READS
161 3,098

4 authors, including:

Ruohan Meng
Nanyang Technological University
19 PUBLICATIONS 410 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ruohan Meng on 24 September 2019.

The user has requested enhancement of the downloaded file.

Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

A Fusion Steganographic Algorithm Based on Faster R-CNN

Ruohan Meng1, 2, Steven G. Rice3, Jin Wang4 and Xingming Sun1, 2, *

Abstract: The aim of information hiding is to embed the secret message in a normal cover
media such as image, video, voice or text, and then the secret message is transmitted
through the transmission of the cover media. The secret message should not be damaged
on the process of the cover media. In order to ensure the invisibility of secret message,
complex texture objects should be chosen for embedding information. In this paper, an
approach which corresponds multiple steganographic algorithms to complex texture
objects was presented for hiding secret message. Firstly, complex texture regions are
selected based on a kind of objects detection algorithm. Secondly, three different
steganographic methods were used to hide secret message into the selected block region.
Experimental results show that the approach enhances the security and robustness.

Keywords: Faster R-CNN, fusion steganography, object detection, CNNs, information

hiding.

1 Introduction
With the rapid development of the internet and information technology, how to protect the
transmission of important information has gained a lot of attentions. Information hiding is
an approach of putting the secret information into a carrier (such as digital images) from
sender to receiver. The receiver extracts and restores the original embedded secret
information through a specific method. The carrier that carrying secret information is called
cover, which has certainly significance in itself. For example, it can be an image, a
document, etc. The carrier adds secret information called stego. In the ideal situation, stego
does not arouse suspicion by attacker in the dissemination process. According to the
different use of information hiding, it is divided into steganography and digital
watermarking, the former is mainly used for the transmission of secret information, and
the latter is mainly used for the protection of intellectual property. According to the
technology of information hiding, it can be divided into a variety of steganographic modes
in spatial domain, transform domain and compressed domain. In the early days, the

1 School of Computer and Software, Nanjing University of Information Science and Technology, Ning Liu
Road, No. 219, Nanjing, 210044, China.
2 Jiangsu Engineering Centre of Network Monitoring, Ning Liu Road, No. 219, Nanjing, 210044, China.

3 Department of Mathematics and Computer Science, Northeastern State University Tahlequah, OK 74464,

USA.
4 School of Computer & Communication Engineering, Changsha University of Science & Technology,

Changsha, 410114, China.

* Corresponding author: Xingming Sun. Email: [email protected].

CMC. doi:10.3970/cmc.2018.055.001 www.techscience.com/cmc

2 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

simplest and most representative method of information hiding in space domain is to hide
the information by using the least significant bit (LSB) of the image or all bit algorithms
of multiple bit planes. However, these algorithms are not robust enough to keep statistical
characteristics, so that attackers can accurately estimate the embedding length according
to the statistical detection methods. The typical detection algorithms are RS (regular and
singular) groups’ method [Fridrich and Goljan (2002)], SPA (sample pair analysis) method
[Dumitrescu, Wu and Wang (2003)] and JPEG compatibility analysis [Fridrich, Goljan and
Du (2001)] and so on. With the continuous development of steganography, the newly
proposed steganography algorithm can maintain more complex image statistical features.
For example, HUGO [Pevný, Filler and Bas (2010)], WOW [Holub and Fridrich (2012)],
SUNIWARD [Holub, Fridrich and Denemark (2014)] and other content adaptive
steganographic algorithms proposed in recent years can automatically embed the secret
information into the rich noisy texture area on the cover image to maintain high-level
statistics.
Deep learning is well known as a revolution in machine learning, especially in the field of
computer vision. In traditional approaches of image feature extraction, the features of SIFT
and BOW are usually used as representation of images. In recent years, the applications of
these features extraction include image retrieval, image forensics and other privacy
protection fields. Xia et al. [Xia, Zhu, Sun et al. (2018); Xia, Xiong, Vasilakos et al. (2017)]
apply image retrieval to cloud computing. Zhou et al. [Zhou, Yang, Chen et al. (2016);
Zhou, Wu, Huang et al. (2017); Zhou, Wu, Yang et al. (2017); Zhou, Wang, Wu et al.
(2017); Cao, Zhou, Sun et al. (2018); Zhou, Mu and Wu (2018)] proposed to apply the
traditional features of the image to coverless information hiding. Yuan et al. [ Yuan, Li,
Wu et al. (2017)] use CNN to detect fingerprint liveness. However, the appearance of
image depth learning features makes the feature extraction more rapid and accurate. In the
field of image classification, deep convolutional neural networks (DCNN) such as
VGGNet [Simonyan and Zisserman (2014)], GoogLeNet [Szegedy, Liu, Jia et al. (2015)]
and AlexNet [Krizhevsky, Sutskever and Hinton (2012)] are performance excellent.
Gurusamy et al. [Gurusamy and Subramaniam (2017)] used a machine learning approach
for MRI Brain Tumor Classification. Based on the achievements above, object detection
has been rapidly developed that detects semantic objects of a class (such as a dog, vehicle,
or person) in digital images and video. For texture-rich target images, the appearance
features of object instances are mainly identified and detected by extracting stable and
abundant feature points and corresponding feature descriptors. In deep learning files, Faster
R-CNN [Ren, Girshick, Girshick et al. (2017)], R-FCN [Dai, Li, He et al. (2016)] and SSD
[Liu, Anguelov, Erhan et al. (2016)] three object detection models that most widely used
currently.
Aiming at information hiding, steganalysis approaches based on deep learning has
appeared one after another. These methods have a very good test for the current
steganography algorithm. For example, Ye et al. [Ye, Ni and Yi (2017)] proposed to
improve the CNN model to detect hidden information has reached 99 % of the detection
rate. At the same time, some researchers turned their attention to how to hide information
based on deep learning, which makes it safer and more robust. Tang et al. [Tang, Tan, Li
et al. (2017)] use generative adversarial network (GAN) to achieve end-to-end information
hiding. Baluja [Baluja (2017)] use neural network to determine the embedding secret
A Fusion Steganographic Algorithm Based on Faster R-CNN 3

information in the location of the image, train an encoder to embed information. Uchida et
al. [Uchida, Nagai, Sakazawa et al. (2017)] embed watermark into the depth neural network
model. Therefore, the combination of deep learning and information hiding has become
the focus of this paper.
The existing methods usually adopt a hidden mode to hide the secret information or
watermarking in the entire image [Xia, Wang, Zhang et al. (2016); Wang, Lian and Shi
(2017); Chen, Zhou, Jeon et al. (2017)]. In order to enhance the security and complexity of
information, it is straightforward to design a fusion hiding strategy that employs different
steganography algorithms to hide information on different areas. In addition, the
complexity of the image is closely related to human visual effects. The more complex the
image, the more information it carries, but the visual information of people does not
increase with the increase of image complexity. Using the effect of human visual
redundancy, the information is hidden in the more complicated area of the image texture,
so as to increase the robustness and anti-detectability of hiding process. Therefore, this
work proposes an approach to detect the more complex area to hide the information by the
method of object detection. We adopt the fusion method of multiple steganographic
algorithms as multiple steganography Algorithms based on ROI (MSA_ROI) to hide the
information.

2 Related works
2.1 Object detection
The aim of object detection is to find the location of all the targets and specify each target
category on a given image or video. It is mainly divided into two tasks, target positioning
and target category detection. In traditional methods, the object detection is mainly use
sliding window framework. The common algorithm is DPM (Deformable Part Model)
[Felzenszwalb, Girshick, Mcallester et al. (2010)]. In face detection, pedestrian detection
and other tasks, DPM have achieved good results. But DPM is relatively complex that the
detection speed is relatively slow. With the development of deep learning, object detection
has entered a new era. Object detection methods related to deep learning can be divided
into two categories, one is based on the regional nomination, such as R-CNN (Region-
based Convolutional Neural Networks) [Girshick, Donahue, Darrell et al. (2014); He,
Zhang, Ren et al. (2015)], SPP-net [He, Zhang, Ren et al. (2015)], Fast R-CNN [Girshick
(2015)], Faster R-CNN, etc. The other is end-to-end approach, Such as YOLO [Redmon,
Divvala, Girshick et al. (2016)] and SSD. At present, Faster R-CNN model is a typical
object detection model based on deep learning. From R-CNN, Fast R-CNN to Faster R-
CNN which used in this paper, the four basic steps of object detection (region proposal,
feature extraction, classification and rectangles refine regression) are finally unified into a
deep network framework to achieve end-to-end object detection.

2.2 Spatial domain steganography

The current steganography methods are mainly divided into two types: spatial domain and
transform domain. In transform domain, main application is JPEG. JPEG is a type of image
compression method that divides an image into several matrices, performs a discrete cosine
variation on each matrix and transforms the specific pixel values on the image completely
4 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

into the frequency domain. Since human eye is sensitive to low frequency but insensitive
to high frequency, all the high frequency information is eliminated so as to achieve the
purpose of image compression. Hidden information will be embedded in the intermediate
frequency area (high-frequency area of anti-attack, low-frequency changes in the region is
easy to be perceived by the human eye), which will be embedded information dispersed
throughout the image. In spatial domain, embedding information is changing pixel values
directly. Adaptive steganography is to automatically select the carrier image which is not
easily found by the attacker based on the content of the cover image features on the region
of interest and embed secret information. Modifying pixels of the image causes less
distortion to the image on the complexity of the rich texture area and the edge area. At the
same time, it is difficult for the attacker to detect secret information.
The steganography that based on the principle of minimizing the embedded distortion
cannot only ensure the minimum distortion rate but also achieves secret communication.
Existing steganography methods that based on the principle of minimizing embedded
distortion include: HUGO, WOW, S-UNIWARD, MVGG [Li, Wang, Huang et al. (2015)]
and so on. The ultimate goal is the same: minimize the distortion function and embed it in
noisy or complex textures region of the cover image.

2.2.1 S-UNIWARD algorithm

S-UNIWARD is a content adaptive steganography based on wavelet transform, which
proposes a general distortion function independent of the embedded domain. The
embedding distortion function of S-UNIWARD as a whole is:
3 n1 n2 Wuv( k ) ( X )  Wuv( k ) (Y )
D( X , Y ) 
k 1 u 1 v 1   Wuv( k ) ( X )
(1)

Here, K (1) , K (2) , K (3) represent horizontal, vertical, diagonal direction of the filter in three
directions, calculated by K (1)  h  g T , K (2)  g  hT , K (3)  g  g T , where h represents a one-
dimensional wavelet decomposition low (high) pass filter. Parameter X represents the
carrier image, the image size is n1  n2 , and parameter Y represents the image after
embedding the message. Parameter Wuv( k ) ( X ),Wuv( k ) (Y ) represent the image of the carrier
image and the encrypted image after wavelet transform. Formula (2) is used to calculate
the wavelet coefficients of pixels in three directions of the original image.
Wuv( k )  K ( k ) * X uv , (1  u  n1 ,1  v  n2 ,1  k  3) (2)
When the wavelet coefficients of pixels (such as texture regions) with complex content
areas are changed, the distortion calculated by formula (1) will be small, indicating that the
region is suitable for hiding information. However, when the pixel wavelet coefficients of
the texture smoothing region are changed, the distortion will be very large, indicating that
secret information should be avoided when embedding these pixels.

2.2.2 HUGO algorithm

HUGO is considered to be one of the most secure steganographic techniques. It defines a
distortion function domain by assigning costs to pixels based on the effect of embedding
some information within a pixel, the space of pixels is condensed into a feature space using
A Fusion Steganographic Algorithm Based on Faster R-CNN 5

a weighted norm function. HUGO algorithm is the use of SPAM steganalysis feature
design distortion cost function. It is considered to be one of the most secure steganographic
techniques. According to the additive distortion function:
n
D( X , Y )   i xi  yi (3)
i 1

Here, the constant 0  i   is a fixed parameter that represents the amount of distortion
that results from a pixel change. When i   , the pixel is the so-called wet pixel, and the
wet pixel does not allow modification during embedding. The minimum expected
distortion function is:
n
Dmin (m, n, p )   pi i (4)
i 1

e  i
Where pi  is the probability that the i-th pixel changes. Parameter
1  e  i
  (  )in1 , (0  i  ) is the set of additive distortion metrics formula (3), where i  {1,..., n} .
Parameter m(0  m  n) is the number of bits to be passed when using binary embedding
operations.

2.2.3 WOW algorithm

WOW (weight acquisition wavelet) is another method of steganography, which depending
on the complexity of the region. It will be covered embedding information into an image.
If one area of the image is more complex than the other, the pixel values in that area will
be modified. WOW steganography algorithm mainly from the perspective design of the
distortion function. The additive distortion function is:
n1 n2
D( X , Y )   ij ( X , Yij ) X ij  Yij
i 1 j 1
(5)
 ij X ij Yij
Where are the costs of changing pixel to .

2.3 Quality assessment

The experimental results in this paper used the following indexes to evaluate the quality of
stego images.

2.3.1 Mean Square Error (MSE)

The following expression is the Mean Square Error (MSE) between the images AI and BI .
Supposing that the pixel value of the bearer image is  AI (i, j),0  j  N 1 and the pixel
value of the corresponding stego image is BI (i, j),0  i  M 1,0  j  N 1 , the error
image is
e(i, j)  AI (i, j)  BI (i, j),0  i  M 1,0  j  N 1 , then the mean square error is
expressed as:
6 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018
M 1 N 1 M 1 N 1
1 1
MSE    e  i, j   MN    AI (i, j )  BI (i, j )
2
2
(6)
MN i 0 j 0 i 0 j 0

Lower is the MSE, higher is the similarity between the images.

2.3.2 Peak Signal to Noise Ratio (PSNR)

The following expression is the Peak Signal to Noise Ratio (PSNR) between the images
AI and BI . Setting AI max  2k  1 , where K represents a number of bits for all pixels, the
PSNR is defined as:
 
AI 2  2
AI max MN 
PSNR  101g  101g  M 1 N 1
max   dB  (7)
 2 
    AI (i, j )  BI (i, j )  
MSE
 i 0 j 0 
In many video sequences and commercial image acquisition applications, often take k = 8.
So, for an 8-bit binary image, AI max  255 , substituting formula (7) into:
 
2552  255 2
MN 
PSNR  101g  101g  M 1 N 1  (dB) (8)
 2 
   
MSE
AI (i , j )  BI (i , j )
 i 0 j 0 
Higher is the PSNR, higher is the similarity between the images.

2.3.3 Structural Similarity Index Measure (SSIM)

The Structure Similarity Index Measure (SSIM) between the images OI and WI which are
of size M×N is given by the following expression. Given two images AI and BI , the
structural similarity of the two images can be found in accordance with the following
formula:
(2 AI  BI  c1 )(2 AIBI  c2 )
SSIM ( AI , BI )  (9)
(  AI
2
  BI2  c1 )( AI
2
  BI2  c2 )
Here,  AI a is the average of AI ,  BI is the average of BI ,  AI2 is the variance of AI ,  BI2 is
the variance of BI , and  AIBI is the covariance of AI and BI . c1  (k1 L)2 , c2  (k2 L) 2 are
constants used to maintain stability. L is the dynamic range of pixel values.
k1  0.01, k2  0.03 . Structural similarities range from -1 to 1. When two images are identical,
the value of SSIM equals one.

3 Proposed method using multiple algorithms based on Faster R-CNN

This section discusses our steganographic scheme, the models we use and the information
each party wishes to conceal or reveal. After laying this theoretical groundwork, we present
experiments supporting our claims. The overall framework for this article is shown in Fig.
1. Firstly, the whole image is input into the CNN model, and then the feature is extracted.
Then, the Proposal is generated through the RPN network, the Proposal is mapped to the
last layer of convolution of CNN, and each proposal is made into a fixed-size feature maps
A Fusion Steganographic Algorithm Based on Faster R-CNN 7

through Rol pooling layer. Finally, use Softmax classification and bounding box regression
to get the part of the carrier we need. Next, the steganographic algorithms are matched for
each of the obtained carrier parts to finally obtain stego images.

3.1 Extract object by Faster R-CNN

In the field of computer vision, object detection mainly solves two problems: the location
of multiple objects on the image and the categories of each object. Faster R-CNN
introduced the region proposal network (RPN) based on Fast R-CNN, replacing the slow
search selective search algorithm. Region proposal uses information such as the texture,
edge, and color in the image to find out the position where the target on the way may appear
beforehand, and can guarantee a higher recall rate with fewer windows selected (a few
hundred or even a few thousand). This greatly reduces the time complexity of follow-up
operations, and obtains the candidate window than the sliding window of higher quality.
In a sense, Faster R-CNN = RPN + Fast R-CNN. Taking into account the target detection
is based on the image texture, edge and determine the target. This paper argues that objects
selected from the Faster R-CNN are more conducive to hiding information than the
background. Therefore, Faster R-CNN is used in the method proposed here. Faster R-
CNN's network model is shown in Fig. 2. Faster R-CNN is mainly divided into four
contents:
(a)
Extract object by Faster R-CNN

Proposals

... Feature
Maps
Proposals ...
Input Region Proposal Classifier
Feature Extraction Rol pooling
Image Network Bounding box

(b)
of Steganographic Algorithm Regions of
Objects

Matching Algorithm Pretreatment

Figure 1: Proposed Steganographic architecture. (a) Target detection structure based on

Faster R-CNN. (b) Steganography algorithm structure for the local area matching
Conv layers: Including the 13 conv layers +13 relu layers +4 pooling layers, used to
extract image features maps. The feature maps are shared for subsequent RPN layers and
full connectivity layers.
Region Proposal Networks: used to generate region proposals. The RPN network first
passes a 3 3 convolutional layer, generating foreground anchors and bounding box
8 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

regression offsets, respectively, and then calculates proposals. Anchors belong to

foreground or background via softmax function.
Roi Pooling: This layer according to feature maps and proposals to extract proposal feature
maps, into the subsequent full connection layer.
Classification: Use proposal feature maps to calculate the type of proposal, and then use
bounding box regression to get the final exact position of the test box.
Conv layers(13), Relu layers(13), Pooling layers(4)

Input ... Feature

Image Map

ROIPooling Proposal Reshape Softmax Reshape 1x1 3x3

1x1

Bbox_pred

Softmax cls_prob

Figure 2: Network model of Faster R-CNN, including convolutional layers, region

proposal network, RoI pooling and classification
The loss function of Faster R-CNN is:
1 1
L({ pi },{ti }) 
Ncls
L
i
cls ( pi , pi* )  
Nreg
p L
i
*
i reg (ti , ti* ) (10)

p i is the predicted probability of anchor i being an object. The ground-truth label:

0 negativelable 
pi*    (11)
1 positive lable 
ti  t x , t y , tw , th  is a vector representing the 4 parameterized coordinates of the bounding
box of the prediction. ti* is the coordinate vector of the ground truth bounding box
corresponding to the positive anchor. Lcls  ti , ti*  is the logarithmic loss target and non-target:
Lcls  pi , pi*    log[ pi* pi  (1  pi* )(1  pi )] (12)
Lreg  ti , ti*  is the regression loss, calculated using Lreg  ti , ti*   R(ti  ti* ) , where R is the
smooth L1 function. pi* Lreg means that only the foreground anchor ( pi*  1) has a regression
loss, and in other cases there is no（pi*  0）. The outputs of cls and reg are composed of
 pi  and ui  , respectively, then normalized by N cls and N reg .
A Fusion Steganographic Algorithm Based on Faster R-CNN 9

3.2 Matching process of steganographic algorithm

From the first part, we get multiple texture complex regions. Next, we need to hide the
information by matching the regions with different steganographic algorithms as shown in
Fig. 3. First of all, the proposed method needs to judge whether there is overlap in the target
area. If there are overlapping parts, the overlap part should be processed. Then the image
should be grayscale. In the order stage, the hash algorithm is used to sort the target area in
each cover image. Finally, the sorted target area and the steganography algorithm is
sequentially matched to complete the concealment of the secret information.

S-UNIWARD
Selection

Regions
of Objects
Grayscale ... WOW Stego

HUGO
Judgement Pretreament Order Matching

Figure 3: Flow chart of steganographic algorithm matching process

3.2.1 Selection of overlapped box

First of all, many of the target frames are overlapped after a target is detected by Faster R-
CNN. Secondly, only one steganographic method can be used to hide the information in
each target frame. Therefore, it needs to preprocess the overlapped box.
As shown in Fig. 4, we take the principle of maximum area, for overlapping target box,
first calculate the area of overlapped box, with a large target box shall prevail, small target
box by removing the remaining large area as a carrier.

(a) (b) (c) (d)

Figure 4: Preprocess for overlapping target box. (a) Target area obtained by Faster R-CNN.
From the graph, we can see that there are overlapping areas in the red border. If the
corresponding steganography algorithm is based on the box, it will cause the overlapping
area to hide the information repeatedly. (b) Removing the overlapped area to get the green
area. Due to the principle of choosing maximum area. (c), the result is shown in the purple
area on the choice of maximum area. The blue and green areas in (d) are the areas of the
final choice

3.2.2 Order of probability scores

10 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

After obtaining the grayscale regional target maps, we apply different steganographic
algorithms to different regions for embedding information. Each target graph obtained by
Faster R-CNN has a probability value of Softmax function. We sort different regions based
on this value. As shown in Fig. 5, there are three boxes in the figure, from left to right
named box 1, box 2, box 3. The probability scores of each box are 0.995, 0.968, and 0.994.
According to descending order, the probability scores are 0.995, 0.994, and 0.0968. The
sorted areas correspond to the boxes 1, 3, and 2, respectively numbered as 1, 2, and 3.

Figure 5: Target areas of the cover image. For different regions have different probability
values, from left to right are 0.995, 0. 0968, 0.994

3.2.3 Match steganographic algorithms

After getting the order of probability scores, we use hash algorithm to match
steganographic algorithms. The method is using division hash algorithm on the number,
taking the remainder of 3. When the remainder is 0, S-UNIWARD algorithm is used. When
the remainder is 1, WOW algorithm is used and when the remainder is 2, we use the HUGO
algorithm. As shown in Fig. 6, the stego image is used by different matching
steganographic algorithms to embed the information in different regions.

Figure 6: The stego image

4 Experiments
All the experiments GPU environment is NVIDIA GTX1080. The experiment training data
set is COCO2014 dataset [Lin, Maire, Belongie et al. (2014)], and the target in the image
is calibrated by exact segmentation. The image includes 91 categories of targets, 328,000
images and 2,500,000 labels. All experiments used the depth learning framework Caffe
[Jia, Shelhamer, Jeff et al. (2014)]. The network used by the image feature extraction
section when training Faster R-CNN is VGG16 [Simonyan and Zisserman (2014)] network.

4.1 Object extraction

A Fusion Steganographic Algorithm Based on Faster R-CNN 11

When we extract the target area, we use the Faster R-CNN model on COCO2014 data set.
When training, the learning rate (base_lr) is set to 0.001, gamma is set to 0.1, momentum
is set to 0.9 and weight decay is set to 0.0005. After 200000 iterations, the network model
file is obtained. When testing, we use this network model file to get the object detection
area box of the image. Fig. 7 is the image obtained after training the test image through the
Faster R-CNN model.

(a) (b) (c) (d) (e)

Figure 7: The first row is the five images from COCO2014 dataset used in testing process.
The second row is the target area after the selection by Faster R-CNN

4.2 Steganographic process

After obtaining the coordinates of the object detection region of the image, three different
steganographic algorithms are used to embed the secret information in the region to obtain
the stego image. The three spatial steganographic algorithms are respectively S-
UNIWARD, HUGO and WOW. According to the division hash, different regions
correspond to different steganography algorithms. Fig. 8 shows the test results after
steganography.

Cover

Residual
map

Stego

Figure 8: The first row shows the test images, the second row represents the residual
images steganography by the proposed method, and the third row shows the stego images
12 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

(a) (b)

(c)
Figure 9: (a), (b) and (c) are line charts of the PSNR values, MSE values and SSIM values
of the proposed method and the three traditional spatial algorithms, respectively

4.3 Analysis the quality of stego

In this section, three different image quality indexes are used to evaluate the quality of the
stego images, which are MSE, PSRN, and SSIM. Through experiments, it is found that the
proposed method is superior to HUGO, S-UNIWARD and WOW steganography in three
indexes of PSNR, MSE and SSIM, indicating the least distortion of the method. As shown
in Tab. 1, the PSNR value of SMSA_ROI is higher than that of HUGO, S-UNIWARD and
WOW. The highest value is 66.6751db, which is higher than 9.24 db in WOW algorithms,
indicating that the distortion of MSA_ROI method is the smallest. The maximum MSE
value of the proposed method is 0.0732, which is lower than the other three methods, and
the minimum can be as low as 0.0140. The last evaluation index SSIM, the four methods
of SSIM value is not much difference, but MSA_ROI is still better than the other three
algorithms. As shown in Fig. 9, the PSNR value of the proposed method is obviously higher
than the three traditional spatial algorithms. The MSE value of the proposed method is
obviously lower than the other three methods, and the lowest is 0.0140. The proposed
method is higher than the three traditional spatial algorithms on the SSIM value, and the
maximum is up to 1.0000.
A Fusion Steganographic Algorithm Based on Faster R-CNN 13

Table 1: Comparison of the proposed algorithm and the single spatial algorithm in image
quality
PSNR MSE
MSA_ROI_A 66.6751 0.0140
HUGO_A 57.8636 0.1063
S-UNIWARD _A 58.4827 0.0922
WOW_A 57.4344 0.1174
MSA_ROI_B 60.1815 0.0624
HUGO_B 58.7124 0.0875
S-UNIWARD _B 59.2457 0.0774
WOW_B 58.3996 0.0940
MSA_ROI_C 60.8032 0.0540
HUGO_C 58.3973 0.0940
S-UNIWARD _C 58.9462 0.0829
WOW_C 58.4119 0.0937
MSA_ROI_HUGO_D 63.0540 0.0322
MSA_ROI_ S-UNIWARD _D 63.5727 0.0268
MSA_ROI_WOW_D 62.9983 0.0326
HUGO_D 58.5103 0.0916
S-UNIWARD _D 59.2435 0.0774
WOW_D 58.1695 0.0991
MSA_ROI_E 59.4835 0.0732
HUGO_E 58.6009 0.0897
S-UNIWARD _E 59.1424 0.0792
WOW_E 58.4468 0.0930

5 Conclusion and future work

This paper has two main contributions. The first one is to combine the object detection
method to select a complex texture region, which is suitable for hiding information. The
second one integrates the existing multiple spatial steganography algorithms into a cover
image. Experiments show that the proposed method is superior to the traditional spatial
steganography algorithm. Future works to further move this research includes the
following aspects. 1. Hide secret message in the foreground completely. 2. Switch to
different object detection methods. 3. Adjust the steganography algorithm adaptively.

Acknowledgement: This work is supported, in part, by the National Natural Science

Foundation of China under grant numbers U1536206, U1405254, 61772283, 61602253,
61672294, 61502242; in part, by the Jiangsu Basic Research Programs-Natural Science
Foundation under grant numbers BK20150925 and BK20151530; in part, by the Priority
14 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

Academic Program Development of Jiangsu Higher Education Institutions (PAPD) fund;

in part, by the Collaborative Innovation Center of Atmospheric Environment and
Equipment Technology (CICAEET) fund, China.

References
Baluja, S. (2017): Hiding images in plain sight: Deep steganography. Advances in Neural
Information Processing Systems, pp. 2066-2076.
Cao, Y.; Zhou, Z.; Sun, X.; Gao, C. (2018): Coverless information hiding based on the
molecular structure images of material. Computers, Materials & Continua, vol. 54, no. 2,
pp. 197-207.
Chen, B.; Zhou, C.; Jeon, B.; Zheng, Y.; Wang, J. (2017): Quaternion discrete fractional
random transform for color image adaptive watermarking. Multimedia Tools and
Application.
Dai, J.; Li, Y.; He, K.; Sun, J. (2016): R-FCN: Object detection via region-based fully
convolutional networks. Advances in Neural Information Processing Systems, pp. 379-387.
Dumitrescu, S.; Wu, X.; Wang, Z. (2003): Detection of LSB steganography via sample
pair analysis. IEEE Transactions on Signal Processing, vol. 51, no. 7, pp. 1995-2007.
Felzenszwalb, P. F.; Girshick, R. B.; Mcallester, D.; Ramanan, D. (2014): Object
detection with discriminatively trained part-based models. IEEE Transactions on Pattern
Analysis & Machine Intelligence, vol. 47, no. 2, pp. 6-7.
Fridrich, J.; Goljan, M. (2012): Practical steganalysis of digital images: state of the art.
Security and Watermarking of Multimedia Contents IV, vol. 4675, no. 1, pp. 1-13.
Fridrich, J.; Goljan, M.; Du, R. (2001): Steganalysis based on JPEG compatibility. Proc
Spie, vol. 4518, pp. 275-280.
Girshick, R.; Donahue, J.; Darrell, T.; Malik, J. (2014): Rich feature hierarchies for
accurate object detection and semantic segmentation. IEEE Conference on Computer
Vision and Pattern Recognition, pp. 580-587.
Girshick, R. (2015): Fast R-CNN. IEEE International Conference on Computer Vision,
pp. 1440-1448.
Gurusamy, R.; Subramaniam, V. (2017): A machine learning approach for MRI brain
tumor classification. Computers, Materials & Continua, vol. 53, no. 2, pp. 91-108.
He, K.; Zhang, X.; Ren, S.; Sun, J. (2015): Spatial pyramid pooling in deep convolutional
networks for visual recognition. IEEE Transactions on Pattern Analysis & Machine
Intelligence, vol. 37, no. 9, pp. 1904-1916.
Holub, V.; Fridrich, J. (2012): Designing steganographic distortion using directional
filters. IEEE International Workshop on Information Forensics and Security, vol. 2, no. 4,
pp. 234-239.
Holub, V.; Fridrich, J.; Denemark, T. (2014): Universal distortion function for
steganography in an arbitrary domain. Eurasip Journal on Information Security, vol. 2014,
no. 1, pp. 1.
A Fusion Steganographic Algorithm Based on Faster R-CNN 15

Jia, Y.; Shelhamer, E.; Donahue, J.; Karayev, S.; Long, J. et al. (2014): Caffe:
Convolutional architecture for fast feature embedding. ACM International Conference on
Multimedia, pp. 675-678.
Krizhevsky, A.; Sutskever, I.; Hinton, G. E. (2012): Imagenet classification with deep
convolutional neural networks. Advances in Neural Information Processing Systems, vol.
60, no. 2, pp. 1097-1105.
Li, B.; Wang, M.; Huang, J.; Li, X. (2015): A new cost function for spatial image
steganography. IEEE International Conference on Image Processing, pp. 4206-4210.
Lin, T. Y.; Maire, M.; Belongie, S.; Hays, J.; Perona, P. et al. (2014): Microsoft COCO:
Common objects in context. European Conference on Computer Vision, vol. 8693, pp.
740-755.
Liu, W.; Anguelov, D.; Erhan, D.; Szegedy, C.; Reed, S. et al. (2016): SSD: Single shot
multiBox detector. European Conference on Computer Vision, pp. 21-37.
Pevný, T.; Filler, T.; Bas, P. (2010): Using high-dimensional image models to perform
highly undetectable steganography. Lecture Notes in Computer Science, vol. 6387, pp.
161-177.
Redmon, J.; Divvala, S.; Girshick, R.; Farhadi, A. (2016): You only look once: unified,
real-time object detection. IEEE Conference on Computer Vision and Pattern Recognition,
IEEE Computer Society, pp. 779-788.
Ren, S.; Girshick, R.; Girshick, R.; Sun, J. (2017): Faster R-CNN: Towards real-time
object detection with region proposal networks. IEEE Transactions on Pattern Analysis &
Machine Intelligence, vol. 39, no. 6, pp. 1137-1149.
Simonyan, K.; Zisserman, A. (2014): Very deep convolutional networks for large-scale
image recognition. Computer Science.
Szegedy, C.; Liu, W.; Jia, Y.; Sermanet, P.; Reed, S. et al. (2015): Going deeper with
convolutions. IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-9.
Tang, W.; Tan, S.; Li, B.; Huang, J. (2017): Automatic steganographic distortion
learning using a generative adversarial network. IEEE Signal Processing Letters, vol. 24,
no. 10, pp. 1547-1551.
Uchida, Y.; Nagai, Y.; Sakazawa, S.; Satoh, S. I. (2017): Embedding watermarks into deep
neural networks. ACM on International Conference on Multimedia Retrieval, pp. 269-277.
Wang, J.; Lian, S.; Shi, Y. (2017): Hybrid multiplicative multi-watermarking in DWT
domain. Multidimensional Systems and Signal Processing, vol. 28, no. 2, pp. 617-636.
Xia, Z.; Wang, X. H.; Zhang, L. G.; Qin, Z.; Sun, X. et al. (2016): A privacy-preserving
and copy-deterrence content-based image retrieval scheme in cloud computing. IEEE
Transactions on Information Forensics and Security, vol. 11, no. 11, pp. 2594-2608.
Ye, J.; Ni, J.; Yi, Y. (2017): Deep learning hierarchical representations for image
steganalysis. IEEE Transactions on Information Forensics & Security, vol. 12, no. 11, pp.
2545-2557.
Yuan, C.; Li, X.; Wu, Q.; Li, J.; Sun, X. (2017): Fingerprint liveness detection from
different fingerprint materials using convolutional neural network and principal component
analysis. Computers, Materials & Continua, vol. 53, no. 4, pp. 357-371.
 16 Copyright © 2018 Tech Science Press CMC, vol.55, no.1, pp.1-16, 2018

Zhou, Z.; Mu, Y.; Wu, Q. (2018): Coverless image steganography using partial-duplicate
image retrieval. Soft Computing.
Zhou, Z.; Wang, Y.; Wu, Q.; Yang, C.; Sun, X. (2017): Effective and efficient global
context verification for image copy detection. IEEE Transactions on Information
Forensics and Security, vol. 12, no. 1, pp. 48-63.
Zhou, Z.; Wu, Q.; Huang, F.; Sun, X. (2017): Fast and accurate near-duplicate image
elimination for visual sensor networks. International Journal of Distributed Sensor
Networks, vol. 13, no. 2, pp. 155014771769417.
Zhou, Z.; Wu, Q.; Yang, C.; Sun, X.; Pan, Z. (2017): Coverless image steganography
based on histograms of oriented gradients-based hashing algorithm. Journal of Internet
Technology, vol. 18, no. 5, pp. 1177-1184.
Zhou, Z.; Yang, C.; Chen, B.; Sun, X.; Liu, Q. et al. (2016): Effective and efficient
image copy detection with resistance to arbitrary rotation. IEICE Transactions on
Information and Systems, vol. E99-D, no. 6, pp. 1531-1540.

View publication stats