A depth iterative illumination estimation network for low-light image enhancement based on retinex theory.

Existing low-light image enhancement techniques face challenges in achieving high visual quality and computational efficiency, as well as in effectively removing noise and adjusting illumination in extremely dark scenes. To address these problems, in this paper, we propose an illumination enhancement network based on Retinex theory for fast and accurate brightening of images in low-illumination scenes. Two learning-based networks are carefully constructed: decomposition network and enhancement network. The decomposition network is responsible for decomposing the low-light input image into the initial reflectance and illumination map. The enhanced network includes two sub-modules: the illumination enhancement module and the reflection denoising module, which are used for efficient brightness enhancement and accurate reflectance. Specially, we have established a cascaded iterative lighting learning process and utilized weight sharing to conduct accurate illumination estimation. Additionally, unsupervised training losses are defined to improve the generalization ability of the model. The proposed illumination enhancement framework enables noise suppression and detail preservation of the final decomposition results. To establish the efficacy and superiority of the model, on the widely applicable LOL dataset, our approach achieves a significant 9.16% increase in PSNR compared to the classical Retinex-Net, and a remarkable enhancement of 19.26% compared to the latest SCI method.

Visual Process Monitoring by Data-Dependent Kernel Discriminant Analysis with t-Distributed Similarities.

Visual process monitoring would provide more directly appreciable and more easily comprehensible information about the process operating status as well as clear depictions of the occurrence path of faults; however, as a more challenging task, it has been sporadically discussed in the research literature on conventional process monitoring. In this paper, the Data-Dependent Kernel Discriminant Analysis (D2K-DA) model is proposed. A special data-dependent kernel function is constructed and learned from the measured data, so that the low-dimensional visualizations are guaranteed, combined with intraclass compactness, interclass separability, local geometry preservation, and global geometry preservation. The new optimization is innovatively designed by exploiting both discriminative information and t-distributed geometric similarities. On the construction of novel indexes for visualization, experiments of visual monitoring tasks on simulated and real-life industrial processes illustrate the merits of the proposed method.

A Multiscale Recursive Attention Gate Federation Method for Multiple Working Conditions Fault Diagnosis.

Federated learning (FL) is an effective method when a single client cannot provide enough samples for multiple condition fault diagnosis of bearings since it can combine the information provided by multiple clients. However, some of the client's working conditions are different; for example, different clients are in different stages of the whole life cycle, and different clients have different loads. At this point, the status of each client is not equal, and the traditional FL approach will lead to some clients' useful information being ignored. The purpose of this paper is to investigate a multiscale recursive FL framework that makes the server more focused on the useful information provided by the clients to ensure the effectiveness of FL. The proposed FL method can build reliable multiple working condition fault diagnosis models due to the increased focus on useful information in the FL process and the full utilization of server information through local multiscale feature fusion. The validity of the proposed method was verified with the Case Western Reserve University benchmark dataset. With less local client training data and complex fault types, the proposed method improves the accuracy of fault diagnosis by 23.21% over the existing FL fault diagnosis.

Design of a High-Order Kalman Filter for State and Measurement of A Class of Nonlinear Systems Based on Kronecker Product Augmented Dimension.

A high-order Kalman filter for full-dimensional variables is proposed for a class of dynamic systems whose state model and measurement model are both nonlinear. The filter requires Taylor expansion of the system equations, and then performs Kronecker product operation on the linear part in the Taylor expansion. Finally, a linear dynamic model is achieved based on the full-dimensional vector formed by the state variables and the high-order dimension expansion variables. After designing the filter, the Kalman filter for the original state variables estimation was selected through the projection operator. The excellent performance of the novel filter is analyzed from the aspects of the information utilization of the state estimation value and the size of the state estimation error covariance matrix. The numerical verification is carried out by computer simulation.

A Video Sequence Face Expression Recognition Method Based on Squeeze-and-Excitation and 3DPCA Network.

Expression recognition is a very important direction for computers to understand human emotions and human-computer interaction. However, for 3D data such as video sequences, the complex structure of traditional convolutional neural networks, which stretch the input 3D data into vectors, not only leads to a dimensional explosion, but also fails to retain structural information in 3D space, simultaneously leading to an increase in computational cost and a lower accuracy rate of expression recognition. This paper proposes a video sequence face expression recognition method based on Squeeze-and-Excitation and 3DPCA Network (SE-3DPCANet). The introduction of a 3DPCA algorithm in the convolution layer directly constructs tensor convolution kernels to extract the dynamic expression features of video sequences from the spatial and temporal dimensions, without weighting the convolution kernels of adjacent frames by shared weights. Squeeze-and-Excitation Network is introduced in the feature encoding layer, to automatically learn the weights of local channel features in the tensor features, thus increasing the representation capability of the model and further improving recognition accuracy. The proposed method is validated on three video face expression datasets. Comparisons were made with other common expression recognition methods, achieving higher recognition rates while significantly reducing the time required for training.

Color Occlusion Face Recognition Method Based on Quaternion Non-Convex Sparse Constraint Mechanism.

As the acquisition and application of color images become more and more extensive, color face recognition technology has also been vigorously developed, especially the recognition methods based on convolutional neural network, which have excellent performance. However, with the increasing depth and complexity of network models, the number of calculated parameters also increases, which means the training of most high-performance models depends on large-scale samples and expensive equipment. Therefore, the key to the current research is to realize a lightweight model while ensuring the recognition accuracy. At present, PCANet, a typical lightweight framework for deep learning, has achieved good results in most of the image recognition tasks, but its recognition accuracy for color face images, especially under occlusion, still needs to be improved. Therefore, a color occlusion face recognition method based on quaternion non-convex sparse constraint mechanism is proposed in this paper. Firstly, a quaternion non-convex sparse principal component analysis network model was constructed based on Lp regularization of strong sparsity. Secondly, the fixed point iteration method and coordinate descent method were established to solve the non-convex optimization problem. Finally, the occlusion recognition performance of the proposed method was verified on Georgia Tech, Color FERET, AR, and LFW-A Color face datasets.

A High-Order Kalman Filter Method for Fusion Estimation of Motion Trajectories of Multi-Robot Formation.

Multi-robot motion and observation generally have nonlinear characteristics; in response to the problem that the existing extended Kalman filter (EKF) algorithm used in robot position estimation only considers first-order expansion and ignores the higher-order information, this paper proposes a multi-robot formation trajectory based on the high-order Kalman filter method. The joint estimation method uses Taylor expansion of the state equation and observation equation and introduces remainder variables on this basis, which effectively improves the estimation accuracy. In addition, the truncation error and rounding error of the filtering algorithm before and after the introduction of remainder variables, respectively, are compared. Our analysis shows that the rounding error is much smaller than the truncation error, and the nonlinear estimation performance is greatly improved.

Fault Detection and Isolation of Non-Gaussian and Nonlinear Processes Based on Statistics Pattern Analysis and the k-Nearest Neighbor Method.

Only low-order information of process data (i.e., mean, variance, and covariance) was considered in the principal component analysis (PCA)-based process monitoring method. Consequently, it cannot deal with continuous processes with strong dynamics, nonlinearity, and non-Gaussianity. To this aim, the statistics pattern analysis (SPA)-based process monitoring method achieves better monitoring results by extracting higher-order statistics (HOS) of the process variables. However, the extracted statistics do not strictly follow a Gaussian distribution, making the estimated control limits in Hotelling-T 2 and squared prediction error (SPE) charts inaccurate, resulting in unsatisfactory monitoring performance. In order to solve this problem, this paper presents a novel process monitoring method using SPA and the k-nearest neighbor algorithm. In the proposed method, first, the statistics of process variables are calculated through SPA. Then, the k-nearest neighbor (kNN) method is used to monitor the extracted statistics. The kNN method only uses the paired distance of samples to perform fault detection. It has no strict requirements for data distribution. Hence, the proposed method can overcome the problems caused by the non-Gaussianity and nonlinearity of statistics. In addition, the potential of the proposed method in early fault detection or safety alarm and fault isolation is explored. The proposed method can isolate which variable or its statistic is faulty. Finally, the numerical examples and Tennessee Eastman benchmark process illustrate the effectiveness of the proposed method.

A Differential Privacy Strategy Based on Local Features of Non-Gaussian Noise in Federated Learning.

As an emerging artificial intelligence technology, federated learning plays a significant role in privacy preservation in machine learning, although its main objective is to prevent peers from peeping data. However, attackers from the outside can steal metadata in transit and through data reconstruction or other techniques to obtain the original data, which poses a great threat to the security of the federated learning system. In this paper, we propose a differential privacy strategy including encryption and decryption methods based on local features of non-Gaussian noise, which aggregates the noisy metadata through a sequential Kalman filter in federated learning scenarios to increase the reliability of the federated learning method. We name the local features of non-Gaussian noise as the non-Gaussian noise fragments. Compared with the traditional methods, the proposed method shows stronger security performance for two reasons. Firstly, non-Gaussian noise fragments contain more complex statistics, making them more difficult for attackers to identify. Secondly, in order to obtain accurate statistical features, attackers must aggregate all of the noise fragments, which is very difficult due to the increasing number of clients. We conduct experiments that demonstrate that the proposed method can greatly enhanced the system's security.

Real-Time Updating High-Order Extended Kalman Filtering Method Based on Fixed-Step Life Prediction for Vehicle Lithium-Ion Batteries.

Lithium-ion batteries have become an important power source in low-carbon transportation energy, and the safe operation and remaining useful life prediction are of great significance. Aiming at the shortcomings of existing methods, such as low prediction accuracy and a short prediction period, this paper proposes a real-time update high-order extended Kalman filter method based on fixed-step life prediction for vehicle lithium batteries based on the principle of combining models and data. First, the state model describing the parameters in the dynamic energy attenuation model is established, and the energy attenuation model is regarded as the observation model of the system to meet the requirements of establishing the Kalman filter. Secondly, the multi-step prediction equation of the state model is established by iterative recursion. At the same time, the multi-step prediction equation between the existing energy output value and the future output value is established based on the multi-dimensional Taylor network (MTN). The multiplicative noise term introduced in the dynamic modeling process is regarded as the hidden variable of the system to meet the requirements of establishing the multi-step linear predictive Kalman filter. Finally, the effectiveness of the new method is verified by digital simulation examples.

Fault Detection of Non-Gaussian and Nonlinear Processes Based on Independent Slow Feature Analysis.

Independent component analysis (ICA) is an excellent latent variables (LVs) extraction method that can maximize the non-Gaussianity between LVs to extract statistically independent latent variables and which has been widely used in multivariate statistical process monitoring (MSPM). The underlying assumption of ICA is that the observation data are composed of linear combinations of LVs that are statistically independent. However, the assumption is invalid because the observation data are always derived from the nonlinear mixture of LVs due to the nonlinear characteristic in industrial processes. Under this circumstance, the ICA-based fault detection is unable to provide accurate detection for specific faults of industrial processes. Since the observation data come from the nonlinear mixing of LVs, this makes the observation data change faster than the intrinsic LVs on the time scale. The temporal slowness can be regarded as an additional criterion in the extraction of LVs. The slow feature analysis (SFA) derived from the temporal slowness has received extensive attention and application in MSPM in recent years. Simultaneously, the temporal slowness is expected to make up for the problem that the LVs extracted by ICA have difficulty accurately describing the characteristics of the process. To solve the above problems, this work proposes to monitor non-Gaussian and nonlinear processes using the independent slow feature analysis (ISFA) that combines statistical independence and temporal slowness in extracting the LVs. When the observation data are composed of a nonlinear mixture of LVs, the extracted LVs of ISFA can describe the characteristics of the processes better than ICA, thereby improving the accuracy of fault detection for the non-Gaussian and nonlinear processes. The superiority of the proposed method is verified by a numerical example design and the Tennessee-Eastman process.

Multilevel Fine Fault Diagnosis Method for Motors Based on Feature Extraction of Fractional Fourier Transform.

Motors are the main driving power for equipment operation, and they are also a major factor to promote the development of the motor and the load it drives and its motor control system toward a low-carbon future, reduce carbon emissions, and improve the industrial economy and social economic efficiency. Due to high-speed, long-period, and heavy-load operation, various faults occur; since the existing integer-order Fourier transform methods have not enough able to detect fractional-order faults and lack robustness, it is difficult to realize the fine diagnosis of motor faults, which reduces the safety and reliability of the motor control system. For this reason, on the basis of the powerful extraction ability of the fractional Fourier transform (FRFT) for micro fault features, especially the extraction ability to fit fractional frequency domain faults, this paper intends to establish a multilevel fine fault diagnosis method for fractional-order or integer-order faults. Firstly, this is accomplished by performing the fractional Fourier transform on the acquired data with faults and feature extraction in the multilevel fractional frequency domain and then optimizing the feature extraction model. Secondly, one further step search method is established to determine the projection direction with the largest fault feature. Thirdly, taking the extracted multilevel fault features as input, a multilevel fine fault diagnosis method based on the SVM model is established. Finally, three typical digital simulation examples and actual operating data collected by the ZHS-2 multifunctional motor test bench with a flexible rotor are employed to verify the effectiveness, robustness, and accuracy of this new method. The main contribution and innovation of this paper are that the fractional Fourier transform method based on time domain and frequency domains is introduced. This method can extract the small fault features in the maximum projection direction of the signal in the fractional domain, but detection with other time-frequency methods is difficult; the extracted multilevel fault features are used as input, and the corresponding fault diagnosis model is established, which can improve the accuracy of fault detection and ensure the safe and reliable operation of industrial equipment.

A New Method of Image Classification Based on Domain Adaptation.

Deep neural networks can learn powerful representations from massive amounts of labeled data; however, their performance is unsatisfactory in the case of large samples and small labels. Transfer learning can bridge between a source domain with rich sample data and a target domain with only a few or zero labeled samples and, thus, complete the transfer of knowledge by aligning the distribution between domains through methods, such as domain adaptation. Previous domain adaptation methods mostly align the features in the feature space of all categories on a global scale. Recently, the method of locally aligning the sub-categories by introducing label information achieved better results. Based on this, we present a deep fuzzy domain adaptation (DFDA) that assigns different weights to samples of the same category in the source and target domains, which enhances the domain adaptive capabilities. Our experiments demonstrate that DFDA can achieve remarkable results on standard domain adaptation datasets.

A Novel Data Sampling Driven Kalman Filter Is Designed by Combining the Characteristic Sampling of UKF and the Random Sampling of EnKF.

In order to improve the performance of the Kalman filter for nonlinear systems, this paper contains the advantages of UKF statistical sampling and EnKF random sampling, respectively, and establishes a new design method of sampling a driven Kalman filter in order to overcome the shortcomings of UKF and EnKF. Firstly, a new sampling mechanism is proposed. Based on sigma sampling with UKF statistical constraints, random sampling similar to EnKF is carried out around each sampling point, so as to obtain a large sample data ensemble that can better describe the characteristics of the system variables to be evaluated. Secondly, by analyzing the spatial distribution characteristics of the obtained large sample ensemble, a sample weight selection and assignment mechanism with the centroid of the data ensemble as the optimization goal are established. Thirdly, a new Kalman filter driven by large data sample ensemble is established. Finally, the effectiveness of the new filter is verified by computer numerical simulation experiments.

A New Method of Deep Convolutional Neural Network Image Classification Based on Knowledge Transfer in Small Label Sample Environment.

The problem of deep learning network image classification when a large number of image samples are obtained in life and with only a small amount of knowledge annotation, is preliminarily solved in this paper. First, a support vector machine expert labeling system is constructed by using a bag-of-words model to extract image features from a small number of labeled samples. The labels of a large number of unlabeled image samples are automatically annotated by using the constructed SVM expert labeling system. Second, a small number of labeled samples and automatically labeled image samples are combined to form an augmented training set. A deep convolutional neural network model is created by using an augmented training set. Knowledge transfer from SVMs trained with a small number of image samples annotated by artificial knowledge to deep neural network classifiers is implemented in this paper. The problem of overfitting in neural network training with small samples is solved. Finally, the public dataset caltech256 is used for experimental verification and mechanism analysis of the performance of the new method.

Design Method of High-Order Kalman Filter for Strong Nonlinear System Based on Kronecker Product Transform.

In this paper, a novel design idea of high-order Kalman filter based on Kronecker product transform is proposed for a class of strong nonlinear stochastic dynamic systems. Firstly, those augmenting systems are modeled with help of the Kronecker product without system noise. Secondly, the augmented system errors are illustratively charactered by Gaussian white noise. Thirdly, at the expanded space a creative high-order Kalman filter is delicately designed, which consists of high-order Taylor expansion, introducing magical intermediate variables, representing linear systems converted from strongly nonlinear systems, designing Kalman filter, etc. The performance of the proposed filter will be much better than one of EKF, because it uses more information than EKF. Finally, its promise is verified through commonly used digital simulation examples.

Rotating Machinery Fault Diagnosis Method by Combining Time-Frequency Domain Features and CNN Knowledge Transfer.

Aiming at the problem of fault diagnosis when there are only a few labeled samples in the large amount of data collected during the operation of rotating machinery, this paper proposes a fault diagnosis method based on knowledge transfer in deep learning. First, we describe the data collected during the operation as a two-dimensional image with both time and frequency-domain characteristics. Second, we transform the trained source domain model into a shallow model suitable for small samples in the target domain, and we train the shallow model with small samples with labels. Third, we input a large number of unlabeled samples into the shallow model, and the output result of the system is regarded as the label of the input sample. Fourth, we combine the original data and the data annotated by the shallow model to train the new deep CNN fault diagnosis model so as to realize the migration of knowledge from the expert system to the deep CNN. The newly built deep CNN model is used for the online fault diagnosis of rotating machinery. The FFCNN-SVM shallow model tagger method proposed in this paper compares the fault diagnosis results with other transfer learning methods at this stage, and its correct rate has been greatly improved. This method provides new ideas for future fault diagnosis under small samples.

Design Method for a Higher Order Extended Kalman Filter Based on Maximum Correlation Entropy and a Taylor Network System.

This paper proposes one new design method for a higher order extended Kalman filter based on combining maximum correlation entropy with a Taylor network system to create a nonlinear random dynamic system with modeling errors and unknown statistical properties. Firstly, the transfer function and measurement function are transformed into a nonlinear random dynamic model with a polynomial form via system identification through the multidimensional Taylor network. Secondly, the higher order polynomials in the transformed state model and measurement model are defined as implicit variables of the system. At the same time, the state model and the measurement model are equivalent to the pseudolinear model based on the combination of the original variable and the hidden variable. Thirdly, higher order hidden variables are treated as additive parameters of the system; then, we establish an extended dimensional linear state model and a measurement model combining state and parameters via the previously used random dynamic model. Finally, as we only know the results of the limited sampling of the random modeling error, we use the combination of the maximum correlation estimator and the Kalman filter to establish a new higher order extended Kalman filter. The effectiveness of the new filter is verified by digital simulation.

Three State Estimation Fusion Methods Based on the Characteristic Function Filtering.

There are three state estimation fusion methods for a class of strong nonlinear measurement systems, based on the characteristic function filter, namely the centralized filter, parallel filter, and sequential filter. Under ideal communication conditions, the centralized filter can obtain the best state estimation accuracy, and the parallel filter can simplify centralized calculation complexity and improve feasibility; in addition, the performance of the sequential filter is very close to that of the centralized filter and far better than that of the parallel filter. However, the sequential filter can tolerate non-ideal conditions, such as delay and packet loss, and the first two filters cannot operate normally online for delay and will be invalid for packet loss. The performance of the three designed fusion filters is illustrated by three typical cases, which are all better than that of the most popular Extended Kalman Filter (EKF) performance.

An EKF-Based Fixed-Point Iterative Filter for Nonlinear Systems.

In this paper, a fixed-point iterative filter developed from the classical extended Kalman filter (EKF) was proposed for general nonlinear systems. As a nonlinear filter developed from EKF, the state estimate was obtained by applying the Kalman filter to the linearized system by discarding the higher-order Taylor series items of the original nonlinear system. In order to reduce the influence of the discarded higher-order Taylor series items and improve the filtering accuracy of the obtained state estimate of the steady-state EKF, a fixed-point function was solved though a nested iterative method, which resulted in a fixed-point iterative filter. The convergence of the fixed-point function is also discussed, which provided the existing conditions of the fixed-point iterative filter. Then, Steffensen's iterative method is presented to accelerate the solution of the fixed-point function. The final simulation is provided to illustrate the feasibility and the effectiveness of the proposed nonlinear filtering method.