A robust digital image watermarking technique using auto encoder based convolutional neural networks
01 Dec 2015-pp 1-6
TL;DR: A digital image watermarking technique using auto-encoder based CNN which is robust to different noises and attacks like salt & pepper, Gaussian and JPEG effect and gives better or on par results with the existing methods.
Abstract: Watermarking plays a very important role in providing authentication, ownership and transmission of secret information. The existing techniques of watermarking in literature are based on either spatial domain or transformation domain. Human brain consists of large number of neurons which are capable of doing paralleling tasking accurately. This resulted in the evaluation of neural network architectures, providing wide range of well connected features representing the input of the network. Convolutional Neural Networks(CNN) which were evolved in early 90s became popular and are being in use in wide range of tasks like classification, detection, recognition, patch matching. In this paper, we propose a digital image watermarking technique using auto-encoder based CNN which is robust to different noises and attacks like salt & pepper, Gaussian and JPEG effect. The proposed method, to the best of our knowledge, is the very first attempt of CNN in the domain of watermarking. We compare the performance of the proposed technique with the existing techniques which are based on spatial domain and transform domain. We show that the proposed method of watermarking using auto encoder based CNN (ACNNWM) gives better or on par results with the existing methods.
Citations
More filters
TL;DR: Experimental and analysis results demonstrate that the proposed concealed attack method has better imperceptibility and attack ability in comparison to the existing watermarking attack methods.
Abstract: While existing watermarking attack methods can disturb the correct extraction of watermark information, the visual quality of watermarked images will be greatly damaged. Therefore, a concealed attack based on generative adversarial network and perceptual losses for robust watermarking is proposed. First, the watermarked image is utilized as the input of generative networks, and its generating target (i.e. attacked watermarked image) is the original image. Inspired by the U-Net network, the generative networks consist of encoder-decoder architecture with skip connection, which can combine the low-level and high-level information to ensure the imperceptibility of the generated image. Next, to further improve the imperceptibility of the generated image, instead of the loss function based on MSE, a perceptual loss based on feature extraction is introduced. In addition, a discriminative network is also introduced to make the appearance and distribution of generated image similar to those of the original image. The addition of the discriminative network can remove watermark information effectively. Extensive experiments are conducted to verify the feasibility of the proposed concealed attack method. Experimental and analysis results demonstrate that the proposed concealed attack method has better imperceptibility and attack ability in comparison to the existing watermarking attack methods.
64 citations
05 Aug 2018
TL;DR: This paper proposes an effective approach for security issue of handwriting documents in spatial domain by making use of watermarking technique and achieves high performance regarding such properties as imperceptibility and robustness against distortions caused by JPEG compression, geometric transformation and print-and-scan process.
Abstract: To prevent falsification of handwriting document images, the methods of forensic document examination are widely used to determine the origin and authenticity of a given document. In this paper, we propose an effective approach for security issue of handwriting documents in spatial domain by making use of watermarking technique. To begin with, the handwritten document is pre-processed by replacing gray level values holding high intensity with the mean value of document content. The document is then transformed into standard form to minimize geometric distortion. Next, fully convolutional networks (FCN) is leveraged to detect document's watermarking regions used for hiding secret information wherein an approach of FCN for document layout segmentation is adjusted to solve the problem of watermarking region detection. Lastly, the data hiding process is conducted by dividing gray level values of each connected object situated within watermarking regions into two sets for carrying one watermark bit. The experiments are performed on various handwritten documents, and our approach achieves high performance regarding such properties as imperceptibility and robustness against distortions caused by JPEG compression, geometric transformation and print-and-scan process.
11 citations
Cites background from "A robust digital image watermarking..."
...Moreover, at the time of carrying out this work, we have found watermarking schemes [16], [17] based on convolutional neural network (CNN) designed for natural images but have...
[...]
01 Aug 2018
TL;DR: This paper proposes a blindly invisible watermarking approach for security matter of general grayscale documents that makes the best use of fully convolutional networks (FCN) to detect stable regions used for hiding secret data.
Abstract: In the literature, the watermarking schemes for document images in spatial domain mainly focus on text content, so they need to be further improved to be possibly applied on general content. In this paper, we propose a blindly invisible watermarking approach for security matter of general grayscale documents. In order to detect stable regions used for hiding secret data, we make the best use of fully convolutional networks (FCN). The FCN for the problem of document structure segmentation is adjusted to solve the problem of watermarking regions detection wherein we consider various types of segmented content regions having the same label. The segmented content regions are then known as watermarking regions. Next, the watermarking pattern is constructed with the aim of detecting potential positions where the watermarking process is carried out. Lastly, the watermarking algorithm is developed by dividing gray level values pertaining to each watermarking pattern into two groups for carrying one watermark bit. The experiments are performed on various document contents, and our approach obtains high performance in terms of imperceptibility, capacity and robustness against distortions caused by JPEG compression, geometric transformation and print-and-scan process.
11 citations
Cites background from "A robust digital image watermarking..."
...More recently, Mun et al. [16] put forward a scheme based on CNN....
[...]
...Besides, at the time of performing this work, we have found some watermarking approaches for natural images using convolutional neural network (CNN) [15], [16] but have not found FCN-based watermarking approaches....
[...]
...The authors divide the host image, watermark into non-overlapping blocks and take advantage of CNN’s weight parameters during training phase for watermarking process....
[...]
...Different from the CNN-based approaches wherein the authors leverage weight parameters of learning framework for watermarking process on the fixed size images, here we train the FCN so that the trained network can be used for generating a salient map describing watermarking regions of an arbitrary sized document....
[...]
TL;DR: This work focuses on developing a watermarking framework for determining whether a received document is genuine or falsified, which is performed by hiding a security feature or secret information within it by constructing hiding patterns used for hiding secret information.
Abstract: Motivated by increasing possibility of the tampering of genuine documents during a transmission over digital channels, we focus on developing a watermarking framework for determining whether a received document is genuine or falsified, which is performed by hiding a security feature or secret information within it. To begin with, the input document is transformed into a standard form to minimize geometric distortion. Fully convolutional network (FCN) is utilized to detect document’s watermarking regions. Next, we construct hiding patterns used for hiding secret information. Modifying pixel values of these patterns for carrying secret bits depends on the edge and corner features of document content and the connectivity of their neighboring pixels. Lastly, the watermarking process is conducted by either changing the center pixel of the hiding patterns or changing the ratio between the number of edge features and the number of corner features of subregions within the watermarking regions. The experiments are performed on various binary documents, and our approach gives competitive performance compared to state-of-the-art approaches.
5 citations
Additional excerpts
...Recently, deep learning is also exploited to develop watermarking system, specifically convolutional neural network (CNN)-based schemes [14,28] and our FCN-based approach [9]....
[...]
20 Sep 2019
TL;DR: This work proposes a robust digital watermarking scheme for securing genuine documents by leveraging generative adversarial networks (GAN), and introduces an algorithm that hides a secret information into the document and produces a watermarked document whose content is minimally distorted in terms of normal observation.
Abstract: Data hiding is an effective technique, compared to pervasive black-and-white code patterns such as barcode and quick response code, which can be used to secure document images against forgery or unauthorized intervention. In this work, we propose a robust digital watermarking scheme for securing genuine documents by leveraging generative adversarial networks (GAN). To begin with, the input document is adjusted to its right form by geometric correction. Next, the generated document is obtained from the input document by using the mentioned networks, and it is regarded as a reference for data hiding and detection. We then introduce an algorithm that hides a secret information into the document and produces a watermarked document whose content is minimally distorted in terms of normal observation. Furthermore, we also present a method that detects the hidden data from the watermarked document by measuring the distance of pixel values between the generated and watermarked document. For improving the security feature, we encode the secret information prior to hiding it by using pseudo random numbers. Lastly, we demonstrate that our approach gives high precision of data detection, and competitive performance compared to state-of-the-art approaches.
4 citations
References
More filters
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.
42,067 citations
TL;DR: An efficient and intuitive algorithm is presented for the design of vector quantizers based either on a known probabilistic model or on a long training sequence of data.
Abstract: An efficient and intuitive algorithm is presented for the design of vector quantizers based either on a known probabilistic model or on a long training sequence of data. The basic properties of the algorithm are discussed and demonstrated by examples. Quite general distortion measures and long blocklengths are allowed, as exemplified by the design of parameter vector quantizers of ten-dimensional vectors arising in Linear Predictive Coded (LPC) speech compression with a complicated distortion measure arising in LPC analysis that does not depend only on the error vector.
7,935 citations
Book•
01 Oct 19984,482 citations
"A robust digital image watermarking..." refers background or methods in this paper
...The series of convolutional and sampling layers learns complex features by weight sharing mechanisms (Convolution Kernel operations and sub-sampling) [1]-[3]....
[...]
...From the literature, it is found that neural network architectures are capable of learning the structures and features of images similar to that of parallel tasking of human brain[1]-[3]....
[...]
...So, we are using Convolutional Neural Network based architecture for auto encoder learning which uses shared weights concept, reducing the need of large storage buffers [1][3] and achieves better PSNR....
[...]
03 Aug 2003
TL;DR: A set of concrete bestpractices that document analysis researchers can use to get good results with neural networks, including a simple "do-it-yourself" implementation of convolution with a flexible architecture suitable for many visual document problems.
Abstract: Neural networks are a powerful technology forclassification of visual inputs arising from documents.However, there is a confusing plethora of different neuralnetwork methods that are used in the literature and inindustry. This paper describes a set of concrete bestpractices that document analysis researchers can use toget good results with neural networks. The mostimportant practice is getting a training set as large aspossible: we expand the training set by adding a newform of distorted data. The next most important practiceis that convolutional neural networks are better suited forvisual document tasks than fully connected networks. Wepropose that a simple "do-it-yourself" implementation ofconvolution with a flexible architecture is suitable formany visual document problems. This simpleconvolutional neural network does not require complexmethods, such as momentum, weight decay, structure-dependentlearning rates, averaging layers, tangent prop,or even finely-tuning the architecture. The end result is avery simple yet general architecture which can yieldstate-of-the-art performance for document analysis. Weillustrate our claims on the MNIST set of English digitimages.
2,783 citations
"A robust digital image watermarking..." refers background or methods in this paper
...The series of convolutional and sampling layers learns complex features by weight sharing mechanisms (Convolution Kernel operations and sub-sampling) [1]-[3]....
[...]
...From the literature, it is found that neural network architectures are capable of learning the structures and features of images similar to that of parallel tasking of human brain[1]-[3]....
[...]
...So, we are using Convolutional Neural Network based architecture for auto encoder learning which uses shared weights concept, reducing the need of large storage buffers [1][3] and achieves better PSNR....
[...]
TL;DR: A trainable feature extractor based on the LeNet5 convolutional neural network architecture is introduced to solve the first problem in a black box scheme without prior knowledge on the data and the results show that the system can outperform both SVMs and Le net5 while providing performances comparable to the best performance on this database.
Abstract: This article focuses on the problems of feature extraction and the recognition of handwritten digits. A trainable feature extractor based on the LeNet5 convolutional neural network architecture is introduced to solve the first problem in a black box scheme without prior knowledge on the data. The classification task is performed by support vector machines to enhance the generalization ability of LeNet5. In order to increase the recognition rate, new training samples are generated by affine transformations and elastic distortions. Experiments are performed on the well-known MNIST database to validate the method and the results show that the system can outperform both SVMs and LeNet5 while providing performances comparable to the best performance on this database. Moreover, an analysis of the errors is conducted to discuss possible means of enhancement and their limitations.
306 citations