Chaotic Sequences Research Articles

Application of Watermarking Technology based on Deep Learning in Face Recognition

Background: Face recognition which belongs to biometric recognition has great application value. Nowadays, face recognition based on deep learning has been widely used in many fields such as internet payment, network login and authentication. However, the face recognition deep learning model are easily replaced and tampered with. Once the models are illegally attacked, it will infringe the intellectual property rights of the model owner and cause economic losses. To deal with these threats, we use watermarking technology to add identity into the face recognition deep learning model. When it is replaced or tampered with, we can prove that the model belongs to us by extracting the watermarks. Objective: In this study, our innovate framework is designed to add watermarks into the face recognition deep learning model as identity, which makes it have features of both trigger sets and data sets. The model will be robust enough to resist common machine learning attacks. With special watermarks, its ownership can be guaranteed. Method: We construct a special watermark trigger set and embed it into the model, which makes it trained without human intervention and annotation. To be flexible for a variety of applications, this scheme uses chaotic sequences to label a watermark trigger set, which guarantees the non-generalization of the watermark. The initial value and parameters used in the method are designed respectively as key to the model. We train 4 models with different number of trigger samples, which is used to study the effect of the number of trigger samples on the model accuracy. Results: We successfully propose a watermarking method for adding identity to the face recognition deep learning model. Watermark extraction rate of the proposed framework is 100%, which means our method can successfully prove ownership of the face recognition deep learning model. In destructive experiments, Models subject to fine-tuning attack still have high face recognition rates which are over 99.00%, and extraction rates of watermarks of each model is 100%. Under overwriting attack, the extraction rates of watermarks of models are less than 25%, models cannot maintain the original performance, which means that watermarks can provide protection until the model loses its ability. The experimental results indicate that the proposed scheme is robust against common machine learning attacks and it prevent the model from being replaced and tempering with. Conclusion: The robustness of the proposed method is capable of resisting machine learning attacks and fine-tuning attacks. It also provides good fidelity, safety, practicality, completeness and effectiveness. With the help of special watermarks, related departments can effectively manage face recognition deep learning models. Besides, it can facilitate the commercialization of intelligent models.

Application of image encryption based improved chaotic sequence complexity algorithm in the area of ubiquitous wireless technologies

Application of image encryption based improved chaotic sequence complexity algorithm in the area of ubiquitous wireless technologies

EFR-CSTP: Encryption for face recognition based on the chaos and semi-tensor product theory

Face recognition is a relatively common method humans use to confirm their identity. Protecting the safe transmission of face images has become a hot issue. This paper proposes an encryption method for face images, EFR-CSTP. The EFR-CSTP is very efficient because it only encrypts the facial in the face image. Firstly, the secret keys are generated by Hash. They are the parameters and the initial values of the two-dimensional logistic tent modular map (2D-LTMM), the chaotic sequence required for encryption by 2D-LTMM. Secondly, facial information is recognized by the histogram of oriented gradients (HOG). Note that the entire algorithm only encrypts this facial image. Finally, the facial image is used as the input of the EFR-CSTP, and the ciphertext image is generated by scrambling, STP combined with the XOR diffusion method. We evaluated our algorithm in the IMDB-WIKI dataset. We also evaluate the EFR-CSTP and other algorithms on the same datasets. The evaluation results show that the EFR-CSTP has excellent performance and efficiency.

Distinguish between Stochastic and Chaotic Signals by a Local Structure-Based Entropy.

As a measure of complexity, information entropy is frequently used to categorize time series, such as machinery failure diagnostics, biological signal identification, etc., and is thought of as a characteristic of dynamic systems. Many entropies, however, are ineffective for multivariate scenarios due to correlations. In this paper, we propose a local structure entropy (LSE) based on the idea of a recurrence network. Given certain tolerance and scales, LSE values can distinguish multivariate chaotic sequences between stochastic signals. Three financial market indices are used to evaluate the proposed LSE. The results show that the LSEFSTE100 and LSES&P500 are higher than LSESZI, which indicates that the European and American stock markets are more sophisticated than the Chinese stock market. Additionally, using decision trees as the classifiers, LSE is employed to detect bearing faults. LSE performs higher on recognition accuracy when compared to permutation entropy.

Parameter identification method of J–A hysteresis model based on improved sparrow search algorithm

Aiming at the weak capacity of the parameter identification methods of Jiles–Atherton (J–A) hysteresis model, such as slow convergence speed and low convergence accuracy, an improved sparrow search algorithm (IM-SSA) is proposed based on mutation crossover operator and T-distribution perturbation strategy in this paper to improve the convergence speed and accuracy of parameter identification method. Firstly, the initial population of the sparrow search algorithm is enriched by Tent chaotic sequence, which expand the search area. Then, the adaptive crossover and mutation operator is introduced into the finders population to enrich the diversity of the finders population and balance the global and local search ability of the algorithm. Secondly, the T-distribution perturbation or differential mutation is used to perturb the population after each iteration according to the individual characteristics, which can avoid the population singularity in the later stage of the algorithm and enhance the ability of jump out of the local optimal value of the algorithm. Finally, the proposed parameter identification method is used to identify the parameters of the J–A hysteresis model, which are used to simulate the magnetic properties of Non-oriented silicon steel sheets with different magnetic induction intensity. The calculation result is compared with the experiment ones. The results proved that the proposed method has faster convergence speed and higher convergence accuracy.

High-efficiency medical image encryption method based on 2D Logistic-Gaussian hyperchaotic map

This paper designs a new medical image encryption scheme based on a new hyperchaotic map, which exhibits excellent security and high efficiency that can be applied to medical images of distinctive presentation formats. Firstly a two-dimensional Logistic-Gaussian hyperchaotic map (2D-LGHM) with a wide range of hyperchaos is constructed, and performance test metrics reveal that it has better ergodicity and unpredictability. Based on 2D-LGHM, a new medical image encryption scheme with permutation and multi-directional pixels substitution processes are established. The chaotic sequence is efficiently used to break the correlation between adjacent pixels and change the global pixel value absolutely. By means of the test of diverse medical images, the results display that the scheme has high robustness and effectiveness that can defense against different security attacks and data loss.

A color image encryption based on chaotic system and PSO-BP neural network and DNA mutation

A color image encryption based on the chaotic system, PSO-BP neural network and DNA mutation is proposed in this paper. Firstly, chaotic characteristics of the non-autonomous laser system are analyzed by phase diagram, Lyapunov exponent, and bifurcation diagram. Secondly, the hash value calculated by SHA-256 algorithm is used to change the initial conditions of chaotic system and generate chaotic sequence. Then, the color image is compressed by the back-propagation neural network based on particle swarm optimization. Based on dynamic encoding and decoding rules, image confusion and diffusion and DNA mutation are designed. Finally, experiments verify that the scheme can compress and encrypt color images, save transmission cost and improve the security performance, which is beneficial to the efficient transmission.

Plaintext-related image encryption scheme without additional plaintext based on 2DCS

In the latest research on image compression and encryption, more and more encryption schemes associate key generation mechanism with plaintext to resist various attacks. This paper presents an image encryption scheme related to plaintext based on 2D compressive sensing(2DCS). It is a plaintext correlation compression and encryption scheme with no additional information. The chaotic sequence generated by Henon map randomly selects some positions in the plaintext image. Double random-phase encoding(DRPE) is applied to the pixels to establish the relationship between the plaintext and the key. The encryption process is divided into three parts. The first part is to map out these key pixels and perform double negative positive transformation(DNPT). In the second part, the plaintext image is sampled, quantified and reconstructed by 2DCS. DNA calculation and diffusion were performed on the reconstructed vector. In the third part, the key pixel matrix and diffusion matrix of DNPT are inversely reconstructed to obtain the cipher image. The proposed encryption mechanism can avoid the transmission of additional information related to the plaintext in the channel, which greatly saves the transmission cost and key storage space. The effectiveness and feasibility of the scheme are verified by various security analyses.

A novel image encryption scheme based on compressive sensing, elliptic curves and a new jerk oscillator with multistability

Compression and encryption are the two main steps generally used to ensure fast and secure transmission of digital images over public communication channels. Among the existing crypto compression approaches, those based on compressive sensing and chaotic systems possess good potential. Unfortunately, the present literature presents some limits including speed issue, security issue and reconstruction problem. These problems are mainly due to the poor performances of the measurement matrix exploited. Note that in some cases these measurement matrices are obtained from chaotic sequences with very poor dynamics. To solve this problem, we proposed in this paper, a novel image encryption scheme based on 2D compressive sensing, elliptic curves and a new septic jerk oscillator with multistability. The proposed scheme adopts the compression before encryption topology to reduce the processing time and is divided into two main parts. The first part concerns image compression using compressive sensing. The input image data is divided into 4 blocks. Then, each block is compressed using 2D compressing sensing based on the measurements matrices generated using chaotic sequences coming from the novel septic jerk system. Next, the compressed blocks are combined to form the compressed image. In the second part, to strengthen the security, elliptic curve theory is used to construct a simple and secure S-box for digital image substitution. This S-box is combined with the septic chaotic solutions to change the pixel’s positions and values of the compressed image. The analysis of simulation test indicates that the proposed cipher has good compression effect, good reconstruction capability even for low compression ratio, is fast and highly secure compared to some recent schemes. As such the proposed work can be exploited to compress and secure data in Wireless Body Area Network (WBAN) applications.

Flexible cascade and parallel operations of discrete memristor

As a special nonlinear component, discrete memristor (DM) can lead to excellent chaos performance in its constructed discrete systems. To enhance the diversity and performance of DM-based systems, cascade, parallel and composite operations for DMs are proposed in this paper. Examples are given to demonstrate the operation process, through which new discrete memristive systems can be constructed with larger parameter space and chaotic region. Complex dynamical behaviors are investigated, and these designed maps are realized on a digital signal processor (DSP) development board. The generated chaotic sequences are further tested by NIST software to show random performance.

IWT and RSA based asymmetric image encryption algorithm

An asymmetric image encryption algorithm based on integer wavelet transformation (IWT) and Rivest-Shamir-Adleman (RSA) algorithm is proposed. Firstly, two plain characteristic parameters (PCP) of the plain image are extracted and two random numbers are chosen. Then, a new parameter transformation model (PTM) is constructed to do nonlinear processing for them, and three cipher characteristic parameters (CCP) are got. After applying RSA operation for CCP (seen as plain messages), three cipher messages are obtained. Secondly, a new initial value obtaining model (IOM) for all plain messages and cipher messages is established, by which initial values of 3D chaotic system are produced. Then, three chaotic sequences can be generated. Thirdly, chaotic sequences are used to confuse the plain image by a way of row-column cycle. Then, IWT operation is carried out and the above chaotic sequences are employed to confuse again the wavelet coefficients. Thereafter, inverse IWT is applied to get the confused image, realizing double confusion operations on both spatial domain and frequency domain. Finally, the confused image is diffused as a whole to get the cipher image. Experiment results explain that the proposed algorithm can realize the encryption in short time, and resist effectively against brute-force attack and noise attack.

Sedimentary–tectonic interaction on the growth sequence architecture within the intraslope basins of deep-water Niger Delta Basin

This paper presents a 3D seismic-based case study from the deep-water Niger Delta Basin to investigate sedimentary–tectonic interaction on growth sequence architecture within the thrust-related intraslope or piggyback basins. Gravitational contraction in the lower continental slope had yielded a series of thrust faults and associated folds in the study area, which formed several piggyback basins. These basins were filled by a suite of growth sequences with varying stratigraphic architecture. Analysis of the 3D seismic data recognized three primary seismic facies types respectively as: convergent, draping and chaotic, which contain seven subtypes. These facies types are combined to form different filling successions for convergent or chaotic growth sequences. The convergent growth sequences mainly occur in the deep section of basin fills during strong gravitational deformation, and always began with convergent-baselapping strata succeeded by convergent-thinning strata, representing pond-to-bypass transition in the ponded-basin accommodation space. The chaotic growth sequences mainly occur in the shallow section of basin fills in response to weak gravitational deformation, and usually began with debris-flow deposits succeeded by channel-levee complexes, reflecting dominant erosion-bypass processes in the slope accommodation space. A dynamic fill-and-spill model considering relationship between episodic sedimentation rate and structural growth rate is proposed to explain the formative mechanisms of growth strata units and associated successions. Interaction between glaciation or deglaciation and sea-level change and gravitational deformation history are suggested to be the factor which resulted in the complex stratal stacking patterns, including progradational or retrogradational stacking patterns within convergent growth sequences, and progradational stacking patterns within chaotic growth sequences.

A New Method for Color Image Encryption Using Chaotic System and DNA Encoding

Image encryption is an essential area of visual cryptography that helps protect images when they are shared over the internet. There are many cryptography algorithms applied for many years in encrypting images. In recent years, techniques from chaotic systems have been combined with DNA coding to help encrypt images and protect them from unauthorized use. In this study, we proposed a new method for image encryption based on a new four-dimensional chaotic system and DNA encoding. The proposed method consists of two stages: the pixel positions are scrambled depending on which chaotic sequences are generated from the new chaotic system; and then, an XOR operation is performed on the image scrambled by using the DNA encoding sequence to get the final encryption image. Analyses such as entropy, NPCR, UACI, and key space are used to evaluate the algorithm's performance. The algorithm attains average entropy of 7.99, which is close to the ideal value of 8 with NPCR and UACI of 99.64 and 33.02, respectively. In addition, the key space is 2627. The simulation results show that the proposed method offers high security, good encryption, and is resistant to many types of attacks

Quaternion with quantum walks for designing a novel color image cryptosystem

Data confidentiality plays a paramount part in our everyday lives. While most modern cryptosystems perform skillfully on digital resources, the rapid development of quantum resources may provide sufficient capability in the near future to hack most current cryptosystems because their construction is mathematical model-based. Although quantum algorithms will not be effective until quantum computers are physically realized and have sufficient capability to outperform digital computation, new cryptosystems are required to be designed using quantum technologies and have the capability to be implemented in both digital and quantum resources besides resisting the probable attacks from those resources. Quantum walk is a computational paradigm that can be implemented as a quantum-inspired paradigm and serves as a chaotic dynamical model employed for developing modern cryptosystems. This work presented a new color image cryptosystem using quaternion, quantum walks, and logistic-sine map. The aim of utilizing quaternion is to increase the key space of the presented cryptosystem, in addition to integrating the sequence originating from the logistic-sine map with the probability distribution originating from quantum walks for generating inseparable chaotic sequences utilized in the permutation and substitution phases of the presented image cryptosystem. Experimental outcomes were performed on 21 color images and the average values of UACI is 33.45969%, NPCR is 99.61070%, correlation is -0.00003, Chi-square is 255.904, global entropy is 7.999837, local entropy is 7.902542, and the encryption speed is 0.725868 Megabytes/second, in which the outcomes demonstrated that the offered encryption technique is both efficient and secure.

War Optimization Method for Image Encryption Algorithm Based on A Chaotic Bit-Plane Decomposition

In this paper, develop image encryption algorithm based on a chaotic bit-plane decomposition and optimization algorithm of a War Optimization Algorithm (WOA). Initially, utilization of SHA-256 hash algorithm for computing the plaintext images hash parameter as initial parameter of the fractional Lorenz hyperchaotic system after the process. Use the chaotic sequence for permuting plaintext image in a bit plane to achieve the scrambled images. After that, block the scrambles image into four sub images of similar size and count the hash parameter of every row of every block through the SHA-256 hash algorithm as the initial parameter of the Sine-Tent Logistic chaotic system. Utilize the achieved chaotic sequence to substitute the images. After that, the four sub-block images to get the last encrypted image and the population can be achieved. At last, utilization information entropy of ciphertext images as the fitness function of WOA. Choose the ciphertext image with the optimal information entropy of ciphertext images as the fitness function of the WOA. Select the ciphertext image with the best information entropy from the population as the optimal encrypted image, and then, return the position value of the best war source meanwhile. The proposed method is implemented in MATLAB and performance is analyzed with performance metrices. The proposed method is compared with the conventional techniques.

A New Chaotic Image Encryption Algorithm Based on Transversals in a Latin Square

In this paper, a new combinatorial structure is introduced for image encryption, which has an excellent encryption effect on security and efficiency. An n-transversal in a Latin square has the function of classifying all the matrix's positions, and it can provide a pair of orthogonal Latin squares. Employing an n-transversal of a Latin square, we can permutate all the pixels of an image group by group for the first time, then use two Latin squares for auxiliary diffusion based on a chaotic sequence, and finally, make use of a pair of orthogonal Latin squares to perform the second scrambling. The whole encryption process is "scrambling-diffusion-scrambling". The experimental results indicated that this algorithm passed various tests and achieved a secure and fast encryption effect, which outperformed many of the latest papers. The final information entropy was very close to 8, and the correlation coefficient was approximately 0. All these tests verified the robustness and practicability of the proposed algorithm.

Image encryption algorithm based on semi-tensor product theory

An image encryption algorithm is proposed by combining chaotic system with semi-tensor product theory. First, the four-way diffusion is performed on the plaintext image with a random interpolation. To further destroy or eliminate the image structure features, the image pixels are split into the octal plane and confused by Josephs scrambling with a variable step. Finally, the ciphertext image is obtained by performing a semi-tensor product operation on an encryption matrix. In this algorithm, the selection of the positions in the chaotic sequence is highly correlated with the plaintext image, which ensures the security of the algorithm without affecting the encryption efficiency.

Research on Multiobjective Optimal Operation Strategy for Wind-Photovoltaic-Hydro Complementary Power System

To address the problems of wind and solar generation volatility and lose of wind and photovoltaic resources, on the basis of the complementary property of wind-solar-water, the topology structure of the wind-solar-water synergy power generation system is constructed. Taking the minimum grid fluctuation index, the minimum wind-photovoltaic-hydro discard rate and the greatest economic effectiveness of the power station as the goal functions and considering various constraints of the wind, photovoltaic, and hydrostation units, a triobjective optimization running model of the wind-photovoltaic-water synergy system is established. Meanwhile, this essay suggests an IMOSSA on the basis of tent chaotic sequence and random wandering strategy to settle the described triobjective optimization issue. Taking Hubei Pankou as an example for simulation analysis, after choosing the best scheme, IMOSSA compared with MOSSA, MOGWO, and NSGA-II, the volatility of sunny days is reduced by 12.39%, 19.5%, and 36.71%, respectively; the wind-photovoltaic abandonment rate is reduced by 11.17%, 22.5%, and 38.03%, respectively, while in the rainy days the volatility is reduced by 8.09%, 18.34%, and 47.03%, respectively; the wind-photovoltaic abandonment rate is reduced by 14.84%, 16.86%, and 40%, respectively. Therefore, it is possible to demonstrate the validity of the proposed three-objective model and the efficiency of the IMOSSA in solving the issue. The efficiency of the optimization operation approach suggested in this research is confirmed by the case study, providing a new idea for the large-scale consumption of new energy in high-proportion hydropower grids.

Color Image Encryption Based on Deep Learning and Block Embedding

Because of the advent of the information age, digital multimedia data are mainly transmitted through the Internet. Image is one of the most popular digital multimedia data. Therefore, this paper proposes a block-based key embedding method and a color image encryption scheme based on deep learning with this method. By using a neural network model to predict the initial chaotic sequence, the key data generated by prediction are encrypted to the color image in layers and blocks. This paper proposed the color image encryption scheme time complexity O n 2 , which is simple and effective. Simulation results show that the proposed scheme exhibits good statistical properties with information entropy mean close to eight and correlation coefficient close to zero. Good robustness to common image attacks like noise addition and cropping. Excellent encryption performance includes enormous key space and low PSNR.

Logistic-ELM: a novel fault diagnosis method for rolling bearings

The fault diagnosis of rolling bearings is a critical technique to realize predictive maintenance for mechanical condition monitoring. In real industrial systems, the main challenges for the fault diagnosis of rolling bearings pertain to the accuracy and real-time requirements. Most existing methods focus on ensuring the accuracy, and the real-time requirement is often neglected. In this paper, considering both requirements, we propose a novel fast fault diagnosis method for rolling bearings, based on extreme learning machine (ELM) and logistic mapping, named logistic-ELM. First, we identify 14 kinds of time-domain features from the original vibration signals according to mechanical vibration principles and adopt the sequential forward selection strategy to select optimal features from them to ensure the basic predictive accuracy and efficiency. Next, we propose the logistic-ELM for fast fault classification, where the biases in ELM are omitted and the random input weights are replaced by the chaotic logistic mapping sequence which involves a higher uncorrelation to obtain more accurate results with fewer hidden neurons. We conduct extensive experiments on the rolling bearing vibration signal dataset of the Case Western Reserve University bearing data centre. The experimental results show that the proposed approach outperforms existing state-of-the-art comparison methods in terms of the predictive accuracy, and the highest accuracies are 100%, 99.71%, 98%, 100%, 100%, and 100%, respectively, in seven separate sub data environments. Moreover, in terms of the runtime cost, the experimental results indicate that the proposed logistic-ELM can predict the fault in 40 ms with a high accuracy, up to 21-1858 times more rapid than existing methods based on support vector machine, convolutional neural network and multi-scale entropy. Other experiments of fault diagnosis of the rolling bearings under four different loads also indicate that the logistic-ELM can adapt to different operation conditions with high efficiency. The relevant code is publicly available at https://github.com/TAN-OpenLab/logistic-ELM .