This work presents an encryption model based on Generative Adversarial Networks (GANs). Encryption of RTF-8 data is realized by dynamically generating decimal numbers that lead to the encryption and decryption of alphabetic strings in integer representation by simple addition rules, the modulus of the dimension of the considered alphabet. The binary numbers for the private dynamic keys correspond to the binary numbers of public reference keys, as defined by a specific GAN configuration. For reversible encryption with a bijective mapping between dynamic and reference keys, as defined by the GAN encryptor, secure text encryption can be achieved by transferring a GAN-encrypted public key along with the encrypted text from a sender to a receiver. Using the technique described above, secure text mail transfer can be realized through component-wise encryption and decryption of text mail strings, with total key sizes of up to 10^8 bits that define random decimal numbers generated by the GAN. From the present model, we assert that encrypted texts can be transmitted more efficiently and securely than from RSA encryption, as long as users of the specific configuration of the GAN encryption model are unaware of the GAN encryptor circuit and configuration, respectively.
In the contemporary era, the abundant availability of health information through internet and mobile technology raises concerns. Safeguarding and maintaining the confidentiality of patients’ medical data becomes paramount when sharing such information with authorized healthcare providers. Although electronic patient records and the internet have facilitated the exchange of medical information among healthcare providers, concerns persist regarding the security of the data. The security of Electronic Health Record Systems (EHRS) can be improved by employing the Cuckoo Search Algorithm (CS), the SHA-256 algorithm, and the Elliptic Curve Cryptography (ECC), as proposed in this study. The suggested approach involves using CS to generate the ECC private key, thereby enhancing the security of data storage in EHR. The study evaluates the proposed design by comparing encoding and decoding times with alternative techniques like ECC-GA-SHA-256. The research findings indicate that the proposed design achieves faster encoding and decoding times, completing 125 and 175 iterations, respectively. Furthermore, the proposed design surpasses other encoding techniques by exhibiting encoding and decoding times that are more than 15.17% faster. These results imply that the proposed design can significantly enhance the security and performance of EHRs. Through the utilization of CS, SHA-256, and ECC, this study presents promising methods for addressing the security challenges associated with EHRs.
Firstly, this paper expounds the conceptual connotation of in-service assessment in the new system, then applies modeling and Simulation in the field of in-service assessment, establishes the conceptual model of in-service assessment and its process, and finally analyzes the application of modeling and simulation in the specific links of in-service assessment.
With the rapid development of various applications of Information Technology, big data are increasingly generated by social network services (SNS) nowadays. The designers and providers of SNS distribute different client applications for PC, Mobile phone, IPTV etc., so that users can obtain related service via mobile or traditional Internet. Good scalability and considerably short time delay are important indices for evaluating social network systems. As a result, investigating and mining the principle of users’ behaviors is an important issue which can guide service providers to establish optimal systems with SNS. On the basis of analyzing the characteristics of social network system, this paper constructed a Stochastic Petri Net (SPN) model for describing the behaviors of three users for SNS. Moreover, the scalability of users’ behaviors of SNS was studied by extending the SPN model of three users to the one of four users. Furthermore, average time delay was chosen as the performance index to evaluate the performance of these two constructed SPN models with Stochastic Petri Net Package (SPNP) 6.0. For different parameters of number of connections, traffic load and buffer size, various trends and numerical results are derived thereby. The methodology of modeling and simulation in this paper can be further used to study the performance of SNS.
Image generation is a hot topic in the academic recently, and has been applied to AI drawing, which can bring Vivid AI paintings without labor costs. In image generation, we represent the image as a random vector, assuming that the images of the natural scene obey an unknown distribution, we hope to estimate its distribution through some observation samples. Especially, with the development of GAN (Generative Adversarial Network), The generator and discriminator improve the model capability through adversarial, the quality of the generated image is also increasing. The image quality generated by the existing GAN based image generation model is so well-paint that it can be passed for genuine one. Based on the brief introduction of the concept of GAN, this paper analyzes the main ideas of image synthesis, studies the representative SOTA GAN based Image synthesis method.
Deep learning related technologies, especially generative adversarial network, are widely used in the fields of face image tampering and forgery. Forensics researchers have proposed a variety of passive forensic and related anti-forensic methods for image tampering and forgery, especially face images, but there is still a lack of overview of anti-forensic methods at this stage. Therefore, this paper will systematically discuss the anti-forensic methods for face image tampering and forgery. Firstly, this paper expounds the relevant background, including the relevant tampering and forgery methods and forensic schemes of face images. The former mainly includes four aspects: conventional processing, fake face generation, face editing and face swapping; The latter is mainly the relevant forensic means based on spatial domain and frequency domain using deep learning technology. Then, this paper divides the existing anti-forensic works into three categories according to their method characteristics, namely hiding operation traces, forgery reconstruction and adversarial attack. Finally, this paper summarizes the limitations and prospects of the existing anti-forensic technologies.
Deep learning is widely used in artificial intelligence fields such as computer vision, natural language recognition, and intelligent robots. With the development of deep learning, people’s expectations for this technology are increasing daily. Enterprises and individuals usually need a lot of computing power to support the practical work of deep learning technology. Many cloud service providers provide and deploy cloud computing environments. However, there are severe risks of privacy leakage when transferring data to cloud service providers and using data for model training, which makes users unable to use deep learning technology in cloud computing environments confidently. This paper mainly reviews the privacy leakage problems that exist when using deep learning, then introduces deep learning algorithms that support privacy protection, compares and looks forward to these algorithms, and summarizes this aspect’s development.
In this paper, a video compressed sensing reconstruction algorithm based on multidimensional reference frames is proposed using the sparse characteristics of video signals in different sparse representation domains. First, the overall structure of the proposed video compressed sensing algorithm is introduced in this paper. The paper adopts a multi-reference frame bidirectional prediction hypothesis optimization algorithm. Then, the paper proposes a reconstruction method for CS frames at the re-decoding end. In addition to using key frames of each GOP reconstructed in the time domain as reference frames for reconstructing CS frames, half-pixel reference frames and scaled reference frames in the pixel domain are also used as CS frames. Reference frames of CS frames are used to obtain higher quality assumptions. The method of obtaining reference frames in the pixel domain is also discussed in detail in this paper. Finally, the reconstruction algorithm proposed in this paper is compared with video compression algorithms in the literature that have better reconstruction results. Experiments show that the algorithm has better performance than the best multi-reference frame video compression sensing algorithm and can effectively improve the quality of slow motion video reconstruction.
In recent years, machine learning has become more and more popular, especially the continuous development of deep learning technology, which has brought great revolutions to many fields. In tasks such as image classification, natural language processing, information hiding, multimedia synthesis, and so on, the performance of deep learning has far exceeded the traditional algorithms. However, researchers found that although deep learning can train an accurate model through a large amount of data to complete various tasks, the model is vulnerable to the example which is modified artificially. This technology is called adversarial attacks, while the examples are called adversarial examples. The existence of adversarial attacks poses a great threat to the security of the neural network. Based on the brief introduction of the concept and causes of adversarial example, this paper analyzes the main ideas of adversarial attacks, studies the representative classical adversarial attack methods and the detection and defense methods.
JPEG (Joint Image Experts Group) is currently the most widely used image format on the Internet. Existing cases show that many tampering operations occur on JPEG images. The basic process of the operation is that the JPEG file is first decompressed, modified in the null field, and then the tampered image is compressed and saved in JPEG format, so that the tampered image may be compressed several times. Therefore, the double compression detection of JPEG images can be an important part for determining whether an image has been tampered with, and the study of double JPEG compression anti-detection can further advance the progress of detection work. In this paper, we mainly review the literature in the field of double JPEG compression detection in recent years with two aspects, namely, the quantization table remains unchanged and the quantization table is inconsistent in the double JPEG compression process, Also, we will introduce some representative methods of double JPEG anti-detection in recent years. Finally, we analyze the problems existing in the field of double JPEG compression and give an outlook on the future development direction.
With the popularization of high-performance electronic imaging equipment and the wide application of digital image editing software, the threshold of digital image editing becomes lower and lower. This makes it easy to trick the human visual system with professionally altered images. These tampered images have brought serious threats to many fields, including personal privacy, news communication, judicial evidence collection, information security and so on. Therefore, the security and reliability of digital information has been increasingly concerned by the international community. In this paper, digital image tamper detection methods are classified according to the clues that they rely on, detection methods based on image content and detection methods based on double JPEG compression traces. This paper analyzes and discusses the important algorithms in several classification methods, and summarizes the problems existing in various methods. Finally, this paper predicts the future development trend of tamper detection.
With the rapid development of the Internet of Things (IoT), all kinds of data are increasing exponentially. Data storage and computing on cloud servers are increasingly restricted by hardware. This has prompted the development of fog computing. Fog computing is to place the calculation and storage of data at the edge of the network, so that the entire Internet of Things system can run more efficiently. The main function of fog computing is to reduce the burden of cloud servers. By placing fog nodes in the IoT network, the data in the IoT devices can be transferred to the fog nodes for storage and calculation. Many of the information collected by IoT devices are malicious traffic, which contains a large number of malicious attacks. Because IoT devices do not have strong computing power and the ability to detect malicious traffic, we need to deploy a system to detect malicious attacks on the fog node. In response to this situation, we propose an intrusion detection system based on distributed ensemble design. The system mainly uses Convolutional Neural Network (CNN) as the first-level learner. In the second level, the random forest will finally classify the prediction results obtained in the first level. This paper uses the UNSW-NB15 dataset to evaluate the performance of the model. Experimental results show that the model has good detection performance for most attacks.
Object detection has been studied for many years. The convolutional neural network has made great progress in the accuracy and speed of object detection. However, due to the low resolution of small objects and the representation of fuzzy features, one of the challenges now is how to effectively detect small objects in images. Existing target detectors for small objects: one is to use high-resolution images as input, the other is to increase the depth of the CNN network, but these two methods will undoubtedly increase the cost of calculation and time-consuming. In this paper, based on the RefineDet network framework, we propose our network structure RF2Det by introducing Receptive Field Block to solve the problem of small object detection, so as to achieve the balance of speed and accuracy. At the same time, we propose a Medium-level Feature Pyramid Networks, which combines appropriate high-level context features with low-level features, so that the network can use the features of both the low-level and the high-level for multi-scale target detection, and the accuracy of the small target detection task based on the low-level features is improved. Extensive experiments on the MS COCO dataset demonstrate that compared to other most advanced methods, our proposed method shows significant performance improvement in the detection of small objects.
Nowadays, machine learning (ML) algorithms cannot succeed without the availability of an enormous amount of training data. The data could contain sensitive information, which needs to be protected. Membership inference attacks attempt to find out whether a target data point is used to train a certain ML model, which results in security and privacy implications. The leakage of membership information can vary from one machine-learning algorithm to another. In this paper, we conduct an empirical study to explore the performance of membership inference attacks against three different machine learning algorithms, namely, K-nearest neighbors, random forest, support vector machine, and logistic regression using three datasets. Our experiments revealed the best machine learning model that can be more immune to privacy attacks. Additionally, we examined the effects of such attacks when varying the dataset size. Based on our observations for the experimental results, we propose a defense mechanism that is less prone to privacy attacks and demonstrate its effectiveness through an empirical evaluation.
Heart rate is an important data reflecting human vital characteristics and an important reference index to describe human physical and mental state. Currently, widely used heart rate measurement devices require direct contact with a person’s skin, which is not suitable for people with burns, delicate skin, newborns and the elderly. Therefore, the research of non-contact heart rate measurement method is of great significance. Based on the basic principle of Photoplethysmography, we use the camera of computer equipment to capture the face image, detect the face region accurately, and detect multiple faces in the image based on multi-target tracking algorithm. Then the region segmentation of the face image is carried out to further realize the signal acquisition of the region of interest. Finally, peak detection, Fourier analysis and wavelet analysis were used to detect the frequency of PPG and ECG signals. The experimental results show that the heart rate information can be quickly and accurately detected even in the case of monitoring multiple face targets.
The widespread acceptance of machine learning, particularly of neural networks leads to great success in many areas, such as recommender systems, medical predictions, and recognition. It is becoming possible for any individual with a personal electronic device and Internet access to complete complex machine learning tasks using cloud servers. However, it must be taken into consideration that the data from clients may be exposed to cloud servers. Recent work to preserve data confidentiality has allowed for the outsourcing of services using homomorphic encryption schemes. But these architectures are based on honest but curious cloud servers, which are unable to tell whether cloud servers have completed the computation delegated to the cloud server. This paper proposes a verifiable neural network framework which focuses on solving the problem of data confidentiality and training integrity in machine learning. Specifically, we first leverage homomorphic encryption and extended diagonal packing method to realize a privacy-preserving neural network model efficiently, it enables the user training over encrypted data, thereby protecting the user’s private data. Then, considering the problem that malicious cloud servers are likely to return a wrong result for saving cost, we also integrate a training validation modular Proof-of-Learning, a strategy for verifying the correctness of computations performed during training. Moreover, we introduce practical byzantine fault tolerance to complete the verification progress without a verifiable center. Finally, we conduct a series of experiments to evaluate the performance of the proposed framework, the results show that our construction supports the verifiable training of PPNN based on HE without introducing much computational cost.
