Image Privacy Research Articles

Image Security Based on Three-Dimensional Chaotic System and Random Dynamic Selection.

Image encryption based on a chaos system can effectively protect the privacy of digital images. It is said that a 3D chaotic system has a larger parameter range, better unpredictability and more complex behavior compared to low-dimension chaotic systems. Motivated by this fact, we propose a new image cryptosystem that makes use of a 3D chaotic system. There are three main steps in our scheme. In the first step, the chaotic system uses the hash value of the plaintext image to generate three sequences. In step two, one of the sequences is used to dynamically select confusion and diffusion methods, where confusion and diffusion have three algorithms, respectively, and will produce 32n (n > 100) combinations for encryption. In step three, the image is divided into hundreds of overlapping subblocks, along with the other two sequences, and each block is encrypted in the confusion and diffusion process. Information entropy, NPCR, UACI results and various security analysis results show that the algorithm has a better security performance than existing, similar algorithms, and can better resist clipping, noise, statistical analysis and other attacks.

A simple chaotic base encryption scheme for securing OFDM-PON communications

Abstract Optical communications and their security have been got a great attention. In this work, a simple, low cost and secure optical system integrating orthogonal frequency division multiplexing-based passive optical network (OFDM-PON) and chaotic theory is proposed. In the proposed system, a novel scheme of quadrature amplitude modulation (QAM) encryption method is adopted. For privacy of image and video encryption, the process of encryption consists of two processes and a 3D chaotic system is employed. The chaotic system is used to generate the encryption sequences. In the first stage, the first generated chaotic sequence is XORed with the incoming data. While in the second stage, the QAM symbol is remapped using the other two generated chaotic sequences. Generally, the proposed system is designed to provide secure data, videos, and images transmission. The proposed system is simulated and evaluated with a transmission of a 10 Gb/s along 50 km single-mode fiber (SMF). The performance of the proposed system showed that there is a performance enhancement in terms of Peak to average power ratio (PAPR) and bit error rate (BER). The BER of unauthorized user is around 0.6147, while the PAPR is reduced to approximately half. The proposed system is simple and doesn’t require complex computations. It requires simple analog multiplexers for its implementation.

Dynamic adversarial domain adaptation based on multikernel maximum mean discrepancy for breast ultrasound image classification

Due to the privacy of breast ultrasound images, it is difficult to obtain a dataset, which results in the lack of a large number of labeled datasets. Additionally, the difference in the feature distribution of datasets is also considered. In this paper, a dynamic adversarial domain adaptive network based on the multi kernel maximum mean discrepancy (MK_DAAN) is proposed. An adaptive layer is added to the model to further align the feature distribution of the datasets in the source and target domains, and the multi kernel maximum mean discrepancy is adopted in the adaptive distance measurement. The dual feature alignment of the adaptive layer and adversarial learning improves the classification performance of the model in breast ultrasound images. The experimental results show that the MK_DAAN model has better classification performance than typical unsupervised domain adaptation (DDC, DAN) and typical adjustment domain adaptation (DAAN, MADA) models.

Speckle-Based Optical Cryptosystem and its Application for Human Face Recognition via Deep Learning.

Face recognition has become ubiquitous for authentication or security purposes. Meanwhile, there are increasing concerns about the privacy of face images, which are sensitive biometric data and should be protected. Software‐based cryptosystems are widely adopted to encrypt face images, but the security level is limited by insufficient digital secret key length or computing power. Hardware‐based optical cryptosystems can generate enormously longer secret keys and enable encryption at light speed, but most reported optical methods, such as double random phase encryption, are less compatible with other systems due to system complexity. In this study, a plain yet highly efficient speckle‐based optical cryptosystem is proposed and implemented. A scattering ground glass is exploited to generate physical secret keys of 17.2 gigabit length and encrypt face images via seemingly random optical speckles at light speed. Face images can then be decrypted from random speckles by a well‐trained decryption neural network, such that face recognition can be realized with up to 98% accuracy. Furthermore, attack analyses are carried out to show the cryptosystem's security. Due to its high security, fast speed, and low cost, the speckle‐based optical cryptosystem is suitable for practical applications and can inspire other high‐security cryptosystems.

PrivacyAlert: A Dataset for Image Privacy Prediction

Image privacy issues have become an important challenge as millions of images are being shared on social networking sites every day. Often due to users' lack of privacy awareness and social pressure, users' posted images reveal sensitive information and may be easily used to their detriment. To address these issues, several recent studies have proposed machine learning models to automatically identify whether an image contains private information. However, progress on this important task has been hampered by the absence of reliable, publicly available, up-to-date datasets. To this end, we introduce PrivacyAlert, a dataset developed from recent images extracted from Flickr and annotated with privacy labels (private or public). Our data collection process is based on state-of-the-art privacy taxonomy and captures a comprehensive set of image types of various sensitivity. We perform a comprehensive analysis of our dataset and report image privacy prediction results using classic and deep learning models to set the ground for future studies. Our dataset is publicly available at: https://doi.org/10.5281/zenodo.6406870.

Adversarial Data Hiding in Digital Images.

In recent studies of generative adversarial networks (GAN), researchers have attempted to combine adversarial perturbation with data hiding in order to protect the privacy and authenticity of the host image simultaneously. However, most of the studied approaches can only achieve adversarial perturbation through a visible watermark; the quality of the host image is low, and the concealment of data hiding cannot be achieved. In this work, we propose a true data hiding method with adversarial effect for generating high-quality covers. Based on GAN, the data hiding area is selected precisely by limiting the modification strength in order to preserve the fidelity of the image. We devise a genetic algorithm that can explore decision boundaries in an artificially constrained search space to improve the attack effect as well as construct aggressive covert adversarial samples by detecting “sensitive pixels” in ordinary samples to place discontinuous perturbations. The results reveal that the stego-image has good visual quality and attack effect. To the best of our knowledge, this is the first attempt to use covert data hiding to generate adversarial samples based on GAN.

A QR code-based user-friendly visual cryptography scheme

Benefiting from the development of the Internet and smart devices, it is now convenient to transmit images anywhere and anytime, which poses a new challenge for image security. The Visual Cryptography Scheme (VCS) is a secret sharing method for protecting an image without a key, the merit of VCS is the human visual system (HVS) can restore the secret image by simply superimposing qualified shares, without any computation. To eliminate noise-like shares in traditional VCS, this paper presents a novel QR code-based expansion-free and meaningful visual cryptography scheme (QEVCS), which generates visually appealing QR codes for transmitting meaningful shares. When distributing on public networks, this scheme does not attract the attention of potential attackers. By limiting the gray-level of a halftoned image, QEVCS both keep the computation-free of visual cryptography and the size of recovery image same as the secret images. The experimental results show the effectiveness of QEVCS when preserving the privacy of images.

Preserving privacy while revealing thumbnail for content-based encrypted image retrieval in the cloud

Owing to the rapid development of cloud services and personal privacy requirements, content-based encrypted image retrieval in the cloud has been increasing. Outsourced images are encrypted into noiselike ones to protect privacy, however, the obtained unrecognized appearance limits their availability. Besides, users have to decrypt all search results to browse, while some of them may not be needed, which undoubtedly wastes bandwidth and computing resources. To cope with this problem, a compromise strategy is proposed that considers the tradeoff between privacy and usability of cipher images. Wherein, a thumbnail preserving encryption (TPE) based on genetic algorithm is proposed. The pixels in the sub-blocks of the plain image are scrambled and diffused at the bit-level through crossover and mutation operators of the genetic algorithm. Moreover, two new operators of Mutation Compensation and Mutation Failure are defined and incorporated into the traditional genetic algorithm to achieve an ideal TPE, that cipher image has the same thumbnail as the original image. Additionally, a color histogram-based retrieval algorithm is introduced to retrieve cipher images using the color information preserved by thumbnails; and to improve retrieval accuracy by using the Bhattacharyya distance. A series of simulations verify the security and effectiveness of our scheme.

IMGTXT: Image to Text Encryption Based on Encoding Pixel Contrasts

Nowadays, when data is exchanged over the internet, the security of data is critical in every element of life. Unauthorized network access is possible due to information transmission. As image usage increased in most communications, image privacy became an issue. Image encryption is one of the methods used to protect images online. In this paper, we proposed a new approach called IMGTXT that converts the image to text by coding the pixel values depending on locations then encrypts them by any trust encryption text algorithm, so that this method provides resistance to a variety of attacks such as histogram attacks and brute force attack. The state of the art of this research is the image is represented as a text and there is no relationship between the cipher-image and the plain image. Although this results in a large data volume. The proposed technique builds and testes on various images with different sizes, the recorded results demonstrate the technique’s efficacy and robustness to resist the brute force attack and statistical cryptanalysis of original and encrypted images.

Privacy‐preserving image retrieval in a distributed environment

Nowadays, several image-based smart services have been widely used in our daily lives, generating many digital images. Since smart devices outsource digital images to the cloud, researchers prefer to select some desired targets from the massive images within the cloud for analysis and improve smart services. Therefore, protective image retrieval on the cloud has attained maximum concentration for privacy-preserving purposes, and the availability assurance of images on the cloud is also a crucial link. Ensuring image security and availability in the cloud environment and precisely preserving retrieval accuracy is comes as a utility-security dilemma while few existing works have explicitly addressed it. Therefore, this paper proposes privacy-preserving image retrieval in the distributed environment based on the combination of image encryption for similarity search and secret image sharing. On the basis of them, we define two-stage encryption. The first-stage encryption algorithm is introduced by modifying Wolfram's reversible cellular automata-based image encryption, which can create a set of processing images to ensure image security and retrieval accuracy. Then, the second-stage encryption algorithm is put forward based on secret image sharing to improve image security and availability. The color histogram could be extracted from the encrypted images for similarity retrieval, and the shadows could be extracted for similar image recovery. Security analysis demonstrates that image privacy and query privacy could be well protected. Moreover, the proposed work achieves more efficient performance for similarity search and similar image recovery compared with some recent works and realizes a reasonable retrieval accuracy on encrypted images for similarity search.

A Color Image Scrambling Method Based on Zigzag Transform and Cross-channel Permutation

It is a challenging task to maintain the privacy and security of color images that are shared over public networks. To address this challenge, image encryption methods can be used to protect the confidentiality of the shared image. As a part of spatial domain-based image encryption, the scrambling process aims to reduce the correlation between adjacent pixels. Furthermore, the correlation between different channels of an RGB color image can be decreased by applying a cross-channel permutation. To achieve both decorrelations, this paper presents a new image scrambling method that is based on zigzag transform and cross-channel permutation. Firstly, a color image’s red, green, and blue channels are obtained, and the zigzag transform is applied to each channel, respectively. The transformed channels are concatenated for further shuffling. A chaotic tent map is employed to create the necessary parameters to perform cross-channel permutation. The cross-channel permutation ensures swapping pixels between different channels. Performance measures of the proposed method show that the correlation coefficients of an RGB image can be significantly reduced. The information entropy of each channel is also greatly increased with this method. The average execution time is found to be less than 0.08 s for a 256 x 256 color image, indicating that the proposed method can be used in image encryption applications.

Chaos blended cellular automata on fractals: the effective way of reconfigurable hardware assisted medical image privacy.

In more recent times data continues to be generated at a very unprecedented scale. This is a result of the pervasive nature of modern-day digitisation. As such, it is absolutely critical that this data only be accessed by the trusted parties concerned in an effort to maintain the privacy of individuals. One particular type data that could severely compromise the identity and privacy of an individual is ‘medical data’. With a focus on medical images, this work proposes a novel ‘fractalized’ chaos-cellular automata encryption scheme, implemented on Cyclone IV EP2C35F672C6 FPGA, resulting in a hardware-based concurrent security solution. The scheme entails three stages of diffusion, which arise from different mechanisms. In tandem with the diffusion process is the “On the Fly” process of confusion governed by a Linear feedback Shift Register (LFSR), all of which in implemented by applying the nature of fractals. The security architecture occupies 16,351 Logic Elements (LEs) with 230 registers on the target FPGA with the power dissipation of 133.39 mW. Further, the encryption achieves near zero correlation with the average entropy of 15.17156 that ensures the statistical properties. In addition, the security framework requires 12.13 ms to encrypt a 256 × 256 × 16 DICOM image which results in the throughput of 86.44 Mbps. The proposed encryption resists the brute force attack and chosen plain text attack by achieving a very large span of keyspace.

A Fast Method for Protecting Users’ Privacy in Image Hash Retrieval System

Effective search engines based on deep neural networks (DNNs) can be used to search for many images, as is the case with the Google Images search engine. However, the illegal use of search engines can lead to serious compromises of privacy. Affected by various factors such as economic interests and service providers, hackers and other malicious parties can steal and tamper with the image data uploaded by users, causing privacy leakage issues in image hash retrieval. Previous work has exploited the adversarial attack to protect the user’s privacy with an approximation strategy in the white-box setting, although this method leads to slow convergence. In this study, we utilized the penalty norm, which sets a strict constraint to quantify the feature of a query image into binary code via the non-convex optimization process. Moreover, we exploited the forward–backward strategy to solve the vanishing gradient caused by the quantization function. We evaluated our method on two widely used datasets and show an attractive performance with high convergence speed. Moreover, compared with other image privacy protection methods, our method shows the best performance in terms of privacy protection and image quality.

Secure medical image encryption with Walsh-Hadamard transform and lightweight cryptography algorithm.

It is important to ensure the privacy and security of the medical images that are produced with electronic health records. Security is ensured by encrypting and transmitting the electronic health records, and privacy is provided according to the integrity of the data and the decryption of data with the user role. Both the security and privacy of medical images are provided with the innovative use of lightweight cryptology (LWC) and Walsh-Hadamard transform (WHT) in this study. Unlike the light cryptology algorithm used in encryption, the hex key in the algorithm is obtained in two parts. The first part is used as the public key and the second part as the user-specific private key. This eliminated the disadvantage of the symmetric encryption algorithm. After the encryption was performed with a two-part hex key, the Walsh-Hadamard transform was applied to the encrypted image. In the Walsh-Hadamard transform, the Hadamard matrix was rotated with certain angles according to the user role. This allowed the encoded medical image to be obtained as a vector. The proposed method was verified with the results of the number of pixel change rates and unified average changing intensity measurement parameters and histogram analysis. The results showed that the method is more successful than the lightweight cryptology method and the proposed methods in the literature to solve security and privacy of the data in medical applications with user roles.

Local Privacy Protection for Sensitive Areas in Multiface Images.

The privacy protection for face images aims to prevent attackers from accurately identifying target persons through face recognition. Inspired by goal-driven reasoning (reverse reasoning), this paper designs a goal-driven algorithm of local privacy protection for sensitive areas in multiface images (face areas) under the interactive framework of face recognition algorithm, regional growth, and differential privacy. The designed algorithm, named privacy protection for sensitive areas (PPSA), is realized in the following manner: Firstly, the multitask cascaded convolutional network (MTCNN) was adopted to recognize the region and landmark of each face. If the landmark overlaps a subgraph divided from the original image, the subgraph will be taken as the seed for regional growth in the face area, following the growth criterion of the fusion similarity measurement mechanism (FSMM). Different from single-face privacy protection, multiface privacy protection needs to deal with an unknown number of faces. Thus, the allocation of the privacy budget ε directly affects the operation effect of the PPSA algorithm. In our scheme, the total privacy budget ε is divided into two parts: ε_1 and ε_2. The former is evenly allocated to each seed, according to the estimated number of faces ρ contained in the image, while the latter is allocated to the other areas that may consume the privacy budget through dichotomization. Unlike the Laplacian (LAP) algorithm, the noise error of the PPSA algorithm will not change with the image size, for the privacy protection is limited to the face area. The results show that the PPSA algorithm meets the requirements ε-Differential privacy, and image classification is realized by using different image privacy protection algorithms in different human face databases. The verification results show that the accuracy of the PPSA algorithm is improved by at least 16.1%, the recall rate is improved by at least 2.3%, and F1-score is improved by at least 15.2%.

MobiDIV: A Privacy-Aware Real-Time Driver Identity Verification on Mobile Phone

As car hire and sharing service is popular in the transportation market, secure driver identity verification is attracting more attention. However, the current verification mechanism focuses on performing authentication operations in the cloud server before drivers get access to the car, which results in potential privacy security issues. In this article, we present a privacy-aware architecture, MobiDIV, which is a client-only scheme, where all sensitive data are processed locally on the driver’s smartphone. To achieve real-time and robust driver identification during the driving life cycle, an efficient face feature extractor is proposed in MobiDIV. Specifically, two three-stream neural networks using the proposed efficient SqueezeNet structure are trained on our synthesized data set for different in-car uncertainties (pose, motion blur, nonalignment and low illumination). During authentication, only an adaptable embedding model is selected and conducted on phone for continuous feature extraction. The anomaly operation monitoring algorithm is then applied to the optical signal generated by phone flash for secure identity reidentification and verification failure message transmission. This allows us to further ensure the privacy of driver facial images without compromising on the real-time identity verification. We perform extensive experiments on various data sets. Compared to most SOTA deep neural networks on real-world open data sets, we achieve similar verification accuracy with fewer parameters and floating-point calculations. On the challenging synthetic test data sets, we even achieve a higher average verification accuracy. To assess the MobiDIV in-depth, the proposed model is integrated in car-sharing platform ICICV-E100 and the obtained results show the feasibility of our system.

Efficient Privacy-Preserving Forensic Method for Camera Model Identification

To address the camera origin identification problem of inquiry images, many forensic methods have been proposed. However, the heavy computational overhead and the potential threat of privacy leakage for inquiry images make many existing forensic methods less applicable. Only a few research works have proposed secure forensic methods to address the aforementioned issues; however, they did not give a detailed analysis for the statistical performance. In this paper, we propose an efficient privacy-preserving forensic method with analytical statistical performance to solve the camera model identification problem efficiently and securely. To preserve the privacy of inquiry images, we propose a hybrid privacy-preserving scheme consisting of two operations: <i>Position Scrambling Encryption</i> to preserve the privacy of image content and <i>Noise Linear-Mapping Processing</i> to preserve the privacy of camera model identity for inquiry images. In the encrypted domain where the proposed privacy-preserving scheme is employed, we first propose a novel statistical noise model, which can accurately characterize an encrypted JPEG inquiry image. Then, a noise model-based detector is designed to identify different camera models. Experimental results verify the feasibility of our proposed method from both privacy-preserving and forensic effectiveness and report that our method outperforms the state-of-the-art secure forensic methods, especially when sample images used to estimate camera fingerprints are insufficient, such as only 2 available images.

Extending Multi-MSB Prediction and Huffman Coding for Reversible Data Hiding in Encrypted HDR Images

Reversible data hiding in encrypted images can simultaneously enhance the image privacy and preserve the message security for the purpose of covert communication and cloud data management. The algorithm extends the multi-MSB prediction and Huffman coding to propose the first reversible data hiding in encrypted HDR images in the Radiance RGBE format. Because of the high similarity among the exponent channel values of neighboring pixels, we directly applied multi-MSB prediction in our proposed preprocessing procedure, yielding considerably increased embedding capacity. We also proposed a novel image encryption method to maintain the characteristics of the images. Subsequently, a distortion-free data hiding algorithm, namely the homogeneity index modification algorithm, was added to further increase embedding capacity. The experimental results demonstrate the feasibility of the proposed algorithm and its ability to increase embedding capacity, embedding rate, and image privacy, support two data hiders, and enable reversibility and separability.

TPE-GAN: Thumbnail Preserving Encryption Based on GAN With Key

To balance the privacy and usability of images in the cloud, Tajik <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">et al.</i> recently designed a thumbnail preserving encryption (TPE) based on sum-preserving encryption, however, multiple iterations make it inefficient. We use CycleGan to simulate randomized unary encoding (RUE) and achieve a more efficient TPE. Thumbnail consistency loss is proposed to guarantee the visual quality of the encrypted image, and the quality of decrypted image is improved via ssim-loss. Besides, a decryption network with a key is retrained such that it can decrypt cipher images in multiple domains, and a binary string is adopted as the key instead of the decryption network parameters to facilitate sharing of images. Simulations verify the effectiveness of the proposed algorithm.

Privacy-Preserving Deep Learning With Learnable Image Encryption on Medical Images

The need for cloud servers for training deep neural network (DNN) models is increasing as more complex architecture designs of DNN models are developed. Nevertheless, cloud servers are considered semi-honest. With great attention to the privacy issues of medical diagnoses using a DNN, previous studies have proposed the idea of learnable image encryption. Though some methods have been presented to partially attack previous encryption schemes, there is still some space for improvement. We proposed a learnable image encryption scheme that is an enhanced version of previous methods and can be used to train a great DNN model and simultaneously keep the privacy of training images. We conducted an experiment on medical datasets from open sources and the result demonstrates the effectiveness of our proposed method in performance and privacy-preserving.