Search

Due to the increasing demand for high-speed motors with high stability, bearingless motors have attracted much attention. Bearingless permanent magnet synchronous motors (BPMSMs) are a kind of motor, which uses suspension force to eliminate the motor bearings. This article proposes a five-phase BPMSM with 10 slots and 8 poles. Based on the principle of suspension force of BPMSM, an analytical expression of the suspension force is derived for the proposed BPMSM. To improve the suspension stability and torque, an optimization method consisting of a response surface model and a multiobjective optimization algorithm is employed. Not only the geometric parameters but also the winding distribution are considered in the optimization process. Finally, the effectiveness and superiority of the optimal design are verified by experiment.

This paper presents a new sensorless control strategy with the fault-tolerant ability for switched reluctance motors (SRMs). Nowadays, many rotor position detection methods are utilized, but most of them increase the system cost due to the additional hardware or complex calculations. To address this problem, various methods of using phase inductance to estimate the rotor position and speed have been proposed. However, the traditional inductance detection method will cause signal distortion due to the presence of interference signals and the unbalanced phase inductance. In particular, the turn-ON angle is advanced. To further improve the stability of the control system, the interference signals are eliminated via establishing the reference time of the inductance characteristic point and the error band in this work. When the position signal disappears, the inductance characteristic points of adjacent phases will be adopted for signal compensation to realize fault-tolerant control. Finally, an SRM drive system is used to verify the effectiveness of the proposed control strategy.

This paper proposes a new position sensorless method for switched reluctance motor (SRM) drive control based on a sliding mode observer using nonlinear Fourier flux linkage model. First, to obtain an accurate mathematical model of SRM for establishment of sliding mode position observer (SMPO), an analytical method based on Fourier polynomial function is introduced to regress the nonlinear relationship between flux linkage, position and current. Then, a new SMPO is designed in detail to estimate the rotor position, including the establishment of SRM model of sliding mode observer, and analysis of estimation error and convergence conditions of SMPO. Finally, position sensorless control of SRM drive based on SMPO is carried out in terms of accuracy and robustness in simulation. Experiments are also given on a dynamic testing platform to verify the effectiveness of the proposed position sensorless scheme.

This paper proposes a model predictive current control (MPCC) compensation method based on virtual voltage vectors for single-phase open-circuit faults of an asymmetric six-phase permanent magnet synchronous hub motor (PMHSM). The proposed strategy adopts the normal vector space to decompose the transformation matrix without reconfiguring the controller topology. By analyzing the difference between the open-circuit phase voltage and the state of health, the disturbance term of the prediction vector in the α-β and x-y subspaces is obtained. 64 voltage vectors are obtained from the switching states of the 6-phase two-level inverter, and then these 64 voltage vectors are appropriately compensated and synthesized into 24 new virtual voltage vectors. The new virtual vectors avoid the adjustment of the MPCC weighting factors and the synthesized virtual vectors can suppress current harmonics. Finally, the experimental results show the effective performance of the method before and after the fault-tolerant operation.

Switched reluctance motors (SRMs) have attracted much attention in industry due to the advantages of low cost, robust structure, high fault tolerance and high torque density. However, several disadvantages like high torque ripples and coil temperatures hinder their industrialization for some applications requiring high dynamic performance, like electric vehicles (EVs). In this paper, a multiobjective and multiphysics design optimization method considering both thermal and electromagnetic performance is presented for a 12/10 SRM. First, the topology of the SRM is introduced and the optimal parameters are defined. Then, the electromagnetic finite element model (FEM) is introduced and the improved transient lumped-parameter thermal model (TLPTM), considering both axial and radial heat transfer for the SRM, is proposed. Second, the objectives and constraints of the optimization are determined. To improve the optimization efficiency, the sequential subspace optimization strategy is employed to find the optimal solution of this high-dimensional design optimization problem. Finally, to validate the effectiveness of the proposed method, both simulation and experimental results are given and discussed. Compared with the initial design, the optimal solution exhibits lower temperature, higher torque, lower torque ripple and less loss.

The exponential growth of social media has spurred an increase in the propagation of hate nowadays. Recent evidence shows that hate speech on social media is detrimental to the mental and physical health of individuals. Thus, there is an emerging need for automated hate speech detection. Automated hate speech detection rests on the intersection between Natural Language Processing (NLP) techniques and machine learning models. An introduction of NLP and its utilities, as well as commonly employed features and classification methods in hate speech detection, are discussed. Hate speech detection in non-English languages is needed to tackle this emergent issue in countries where multiple languages are used. Hence, an overview of the current literature on hate speech detection in non-English languages are covered too. Challenges in the field of hate speech detection are explored and the importance of standardized methodologies for building corpora and data sets are emphasized.

This article presents a sensorless control for the permanent-magnet linear synchronous motor (PMLSM) under the energy-feeding mode in vehicle regenerative suspension. The proposed method is based on a multiple-iterative search algorithm to obtain a limited quantity of mover position signals (electrical angle) precisely, which are utilized to observe the $d$ -axis back electromotive force (EMF). Then, the optimal mover position signal selected from the estimated mover position is restrained by the cost function applied in the multiple-iteration search algorithm. The accuracy of the mover position signal increases geometrically in the effect of an increase in the iterations. The proposed search algorithm is superior to the existing search algorithm in the terms of the computational burden. Different from the conventional phase-locked loop (PLL), the new strategy removes the need for a proportional–integral controller. The proposed control method and the conventional control method are experimentally tested under different conditions. Results prove that the proposed PLL has the effectiveness to realize the estimated parameters corresponding to the actual parameters.

This paper focus on a state feedback controller (SFC)-based optimal control scheme for surface-mounted permanent-magnet synchronous motor (SPMSM) with auto-tuning of controller built on seeker optimization algorithm (SOA). First, based on the nonlinear state-space model of SPMSM, voltage feedforward compensation is used to design a linear SFC. Then in order to guarantee the steady performance in speed and current, integral models considering the errors of rotor speed and current response in d-axis are added in the state space model of SPMSM. Furthermore, by statically decoupling the load torque in the state equation, feedforward compensation is implemented on the load torque to improve the dynamic performance of the controller. The load torque is estimated by using disturbance observer with reasonable parameter selection. Then, with the consideration of the search capacity of seeker optimization algorithm (SOA), it is adopted to acquire matrix coefficient of the presented controller. Furthermore, in order to suppress the speed overshoot, a penalty term is introduced to the fitness index. The performance of the proposed method has been validated experimentally and compared with the conventional method under different conditions.

This paper represents modified model reference adaptive system (MRAS) method for speed sensorless control of interior permanent magnet synchronous motor (IPMSM), which can estimate rotor speed and position information. In order to suppress the adverse effect of parameter variation on control performance, the stator resistance and the permanent magnet flux linkage are estimated and continuously updated in reference and adjustable model. Then, proportional-integral (PI) controller parameters of speed adaptive law obtained by model reference adaptive system are optimized by grey wolf optimization (GWO) algorithm. In order to get the smallest possible speed following error and reference current error, the objective function is designed with discretized rotor speed error and current error as variables. The simulation results show the better performance of rotor speed estimation is obtained with GWO algorithm.

This paper studies an online iterative algorithm for solving discrete-time multi-agent dynamic graphical games with input constraints. In order to obtain the optimal strategy of each agent, it is necessary to solve a set of coupled Hamilton-Jacobi-Bellman (HJB) equations. It is very difficult to solve HJB equations by the traditional method. The relevant game problem will become more complex if the control input of each agent in the dynamic graphical game is constrained. In this paper, an online iterative algorithm is proposed to find the online solution to dynamic graphical game without the need for drift dynamics of agents. Actually, this algorithm is to find the optimal solution of Bellman equations online. This solution employs a distributed policy iteration process, using only the local information available to each agent. It can be proved that under certain conditions, when each agent updates its own strategy simultaneously, the whole multi-agent system will reach Nash equilibrium. In the process of algorithm implementation, for each agent, two layers of neural networks are used to fit the value function and control strategy, respectively. Finally, a simulation example is given to show the effectiveness of our method.

To improve the ability of the sine cosine algorithm (SCA) in the exploitation process, an improved symmetric SCA with adaptive probability selection (SSCA-APS), is proposed. The search process of this algorithm is divided into early and late stages. In the early stage, the operators of the traditional SCA algorithm continue to be used. In the late stage, three improvements were applied. Firstly, the symmetric sine and cosine operators are proposed. The adaptive probability selection strategy is adopted to integrate original sine and cosine operators and symmetric sine and cosine operators for dynamically adjusting the step size of the search range. Furthermore, to prevent the population from falling into local optimization, Gaussian perturbation is used to mutate the globally optimal individuals of the current generation. In addition, the information of two randomly selected individuals and the globally optimal individual is integrated by quadratic interpolation to maintain population diversity and produce a new individual. 23 test functions were used to verify the performance of the proposed algorithm. The simulation results indicate that the performance of the SSCA-APS algorithm has competitiveness when it is compared with classical SCA and some state-of-the-art SCA variants.

Sleep-stage classification is essential for sleep research. Various automatic judgment programs including deep learning algorithms using artificial intelligence (AI) have been developed, but with limitations in data format compatibility, human interpretability, cost, and technical requirements. We developed a novel program called GI-SleepNet, generative adversarial network (GAN)-assisted image-based sleep staging for mice that is accurate, versatile, compact, and easy to use. In this program, electroencephalogram and electromyography data are first visualized as images and then classified into three stages (wake, NREM, and REM) by a supervised image learning algorithm. To increase the accuracy, we adopted GAN and artificially generated fake REM sleep data to equalize the number of stages. This resulted in improved accuracy, and as few as one mouse data yielded significant accuracy. Because of its image-based nature, it is easy to apply to data of different formats, of different species of animals, and even outside of sleep research. Image data can be easily understood by humans, thus especially confirmation by experts is easy when there are some anomalies of prediction. Because deep learning of images is one of the leading fields in AI, numerous algorithms are also available.

Due to the increasing use of prefabricated building blocks in construction projects, on-site storages with limited space need to be utilized more efficiently to start the construction activities relating to the blocks on time. Especially when the construction schedule is disrupted by an unexpected event, it needs to be revised by coordinating the delivery times of prefabricated blocks from suppliers, subject to the limited storage space, for minimizing the impact on the project duration. In this paper, a coordination-based reactive scheduling procedure is proposed for a reactive multimode resource-constrained project scheduling problem with limited on-site storage space and its effectiveness is validated through computational experiments. The authors expect that the proposed procedure is also applicable to engineer-to-order (ETO) and make-to-order (MTO) manufacturing systems with limited on-site storage space for raw materials.

According to the deformation mechanism of parts in the process of UV curing rapid prototyping, the equivalent mechanical model of layer by layer accumulation was established, and the warpage of multi-layer plate parts was simulated by ABAQUS software. The simulation results are in good agreement with the experimental results, which shows that the simulation model can simulate the growth and deformation of the parts accurately. It can provide theoretical support and quantitative basis for improving the manufacturing scheme of parts, such as the selection of growth direction and the addition of support. It provides theoretical support and quantitative basis for improving the process and improving the precision of forming parts.

This paper discusses the developments of employee attendance system via face detection and facial recognition, using the enhanced featured supervised learning technique. The main goal of the proposed system, FaceAuth is to uniquely identify a person without the use of Internet connection and cloud services. The developed system can identify the face of a user using the k-nearest neighbors (k-NN) algorithm for both training and testing phases. Then, a suitable distance threshold is configured to exclude the unregistered person from the prediction. Meanwhile, it marks the attendance of individual, whenever the person's face has trained within five seconds. For 40 participants in this experiment, the developed system can achieve 92.5% in accuracy, without false positive error occur.

In this paper, we present graphics rendering instruction extension to RISC-V ISA. In order to achieve high performance at low power consumption, the fragment rendering process is designed to be fixed-pipeline. The proposed system on a chip (SOC) is fabricated into a 3.5 × 3.5 mm2 chip using 110-nm CMOS technology. The RISC-V core with programmable vertex engine consumes 1.10 mm2 and runs at 102 MHz. The fragment engine consumes 7.44 mm2 and runs at 200 MHz. Experiment Results show that the whole chip only consumes 20 mW at full speed, and the performance per milliwatt is over 200 times more than modern MCU chip with GPU core.

This study deals with a multi-level job scheduling problem in a single machine under processing time uncertainty. For the objectives of minimizing the total weighted tardiness of orders and the schedule deviation of the realized schedule from the baseline schedule, two surrogate measures for each of the two objectives are proposed to cope with the processing time uncertainty in the baseline scheduling phase. For the suggested surrogate measures, several optimal solution properties are developed to reduce the size of the solution space. Then, by using the solution properties a heuristic algorithm with two different order prioritization rules and a metaheuristic algorithm based on the discrete particle swarm optimization are developed for the order sequencing problem. Also, an iterative job sequencing algorithm is developed for the job sequencing problem that can be used for a given order sequence. From a series of computational experiments with a number of randomly generated problem instances with multiple bills of materials and normally distributed processing times of items, it is identified that the suggested metaheuristic algorithm with or without the iterative job sequencing algorithm shows good performance compared with the heuristic algorithm.

An improved global stereo matching algorithm is implemented on a single FPGA for real-time applications. Stereo matching is widely used in stereo vision systems, i.e. objects detection and autonomous vehicles. Global algorithms have much more accurate results than local algorithms, but global algorithms are not implemented on FPGA since they rely over high-end hardware resources. In this implementation the stereo pairs are divided into blocks, the hardware resources are reduced by processing one block once. The hardware implementation is based on a Xilinx Kintex 7 FPGA. Experiment results show that the proposed implementation has an accurate result for the Middlebury benchmarks and 30 frames per second (fps) @1920×1680 is achieved.

Combined with aerodynamic calculation and accelerometer-based method for estimating the angle of attack of the guided munitions, this paper studies the angle of attack identification of a miniature guided munition. Through the establishment of the aerodynamic model, the aerodynamic and balanced angle of attack data under the typical flight conditions of the guided munitions are calculated. Based on the aerodynamic characteristics, this paper establishes an angle of attack identification model and proposes an IMU layout scheme. Combined with the test data, the angle of attack identification results of the guided munitions were calculated. The comparison with the simulation results shows that the angle of attack identification based on flight test data is consistent with the aerodynamic simulation results, which proves that the proposed angle of attack identification method is reasonable and feasible.

With the acceleration of social rhythm and the exacerbation of psychological problems caused by stress, psychological relief and adjuvant therapy based on relaxation and decompression have become the focus of research. Different from the traditional psychological relaxation therapy methods that using images, audios and videos, VR-based relaxation therapy is a promising way as the interaction of the VR system and dynamics of virtual environment affect the effectiveness of relax through more immersive and authentic audio-visual scenes. It makes the patient's emotion adjust to the normal state by generating a simulated environment in a VR system to comfort the subject and relieve subject mental stress. This paper presents a novel VR relaxation scenes design and evaluation framework where subject sensitive tourism VR videos were selected and combined with alpha wave induced background music. The relaxation effectiveness of VR scenes were evaluated by SAM scale, relaxation questionnaire and frontal EEG signals. A number of frontal EEG features from various analysis domains were explored to obtain the best stress-relevant features and to establish a stress-frontal EEG correlation model for stress evaluation, including time, frequency, entropy, geometric analysis, subband spectra, multiscale entropy, etc. 13 students aged between 20 and 30 were recruited in the evaluation experiment of 6 newly designed VR relaxation scenes. During which subjects frontal EEG signals were recorded as well as the SAM scale and questionnaire, the features that demonstrate an significant difference before and after watching VR scenes were identified, such as Energy, Energy ratio, the mean square root of frequency, frequency domain symmetry, Energy entropy and Shannon entropywhere frequency bands of delta, theta, low alpha and high alpha were considered. Finally, the features that demonstrate an significant difference before and after watching VR scenes were identified. The best stress-relevant features were selected to establish a stress-frontal EEG correlation model for stress evaluation which could be used to validate the effectiveness of VR relaxation scenes.

Background Pressure sensors have been used for sleeping posture detection, which meet privacy requirements. Most of the existing techniques for sleeping posture recognition used force-sensitive resistor (FSR) sensors. However, lower limbs cannot be recognized accurately unless thousands of sensors are deployed on the bedsheet. Method We designed a sleeping posture recognition scheme in which FSR sensors were deployed on the upper part of the bedsheet to record the pressure distribution of the upper body. In addition, an infrared array sensor was deployed to collect data for the lower body. Posture recognition was performed using a fuzzy c-means clustering algorithm. Six types of sleeping body posture were recognized from the combination of the upper and lower body postures. Results The experimental results showed that the proposed method achieved an accuracy of above 88%. Moreover, the proposed scheme is cost-efficient and easy to deploy. Conclusions The proposed sleeping posture recognition system can be used for pressure ulcer prevention and sleep quality assessment. Compared to wearable sensors and cameras, FSR sensors and infrared array sensors are unobstructed and meet privacy requirements. Moreover, the proposed method provides a cost-effective solution for the recognition of sleeping posture.

The present study focuses on the fault diagnosis of mechanical systems. Mechanical systems are considered with interconnected components that work together to achieve a common function or purpose. On the one hand, the fault diagnosis result is affected by strong dependence between each component. One the other hand, diagnostic results may be different at different time slices because of the performance degradation of components when the same fault symptoms are given. To deal with these problems in diagnosis, a dynamic Bayesian network (DBN) model is proposed. First, series and parallel systems are converted to a Bayesian network. And the relationship between components and reliability of the system is expressed by the Bayesian network. Then, the dynamic Bayesian network is established to model the dynamic degradation of components in a system under additional information by using the wear data. The parameters of the model are estimated by historical data. Finally, a case is investigated to verify the proposed model in this study. Fault diagnosis is conducted through a backward analysis of the DBN model proposed, and the weakest component is identified. The dynamic probabilities of the mechanical system are obtained through forwarding analysis of the DBN model.

Converting geographic features (e.g., place names) in map images into a vector format is the first step for incorporating cartographic information into a geographic information system (GIS). With the advancement in computational power and algorithm design, map processing systems have been considerably improved over the last decade. However, the fundamental map processing techniques such as color image segmentation, (map) layer separation, and object recognition are sensitive to minor variations in graphical properties of the input image (e.g., scanning resolution). As a result, most map processing results would not meet user expectations if the user does not "properly" scan the map of interest, pre-process the map image (e.g., using compression or not), and train the processing system, accordingly. These issues could slow down the further advancement of map processing techniques as such unsuccessful attempts create a discouraged user community, and less sophisticated tools would be perceived as more viable solutions. Thus, it is important to understand what kinds of maps are suitable for automatic map processing and what types of results and process-related errors can be expected. In this paper, we shed light on these questions by using a typical map processing task, text recognition, to discuss a number of map instances that vary in suitability for automatic processing. We also present an extensive experiment on a diverse set of scanned historical maps to provide measures of baseline performance of a standard text recognition tool under varying map conditions (graphical quality) and text representations (that can vary even within the same map sheet). Our experimental results help the user understand what to expect when a fully or semi-automatic map processing system is used to process a scanned map with certain (varying) graphical properties and complexities in map content. We discuss the impact of graphical quality on automatic map processing.We include a comprehensive experiment covering a wide range of map products.We report our findings on the tested map products under varying image resolutions.

In embedded graphics systems, the graphics processing unit (GPU) also consumes significant amount of frame buffer memory bandwidth. Frame buffer compression is widely adopted to alleviate both memory bandwidth and power consumption issues for the display controller, but rarely has it been applied to addressing GPU’s consumption. This paper proposes a real-time fixed-ratio frame buffer compression technique for RISC-V processor-based embedded graphics systems. The advantage of the proposed method is that the fixed-ratio compressed frame buffer can be directly adopted as an input texture by GPU. The proposed architecture (called an FBC coprocessor) is a hardware extension to the RISC-V microprocessor that supports frame buffer memory bandwidth reduction. The results show that the coprocessor consumes only 1% additional silicon space of the whole system, while reducing bandwidth consumption by 72.64%. A prototype system-on-a-chip indicates that the proposed FBC coprocessor can reduce GPU power consumption by up to 12.7% for an example automotive application.

This paper presents a vision-assisted human motion analysis method that can be used to recognize in-bed preparatory motions for bed-exit. In this work, a 3D convolutional neural network was applied to predict sequential motions from ultra-low resolution (32 × 24 pixels) clips composed of depth images, which meets both privacy and computational efficiency requirements. The experimental results showed that the system effectively recognized key motions of bed-exit before bed-leaving. Moreover, these motions were applied to bed-exit prediction in several cases of disturbance and for beds of different structures, and the system accurately predicted bed-exit.

Precision load control is an important part of UHV AC-DC grid system protection, and it can also be used as a precise switching device for distributed energy, micro-grid and other systems. Through the classification, grading and sub-regional management of load resources, the real-time precise control of the interrupted load of the classified users can be realized by the direct dispatching of the dispatching, which can avoid a large number of overall pull-off of the substation or the line, and reduce the social impact of the grid fault. At the lowest level, improve the defense capability of large power grids. In order to ensure the stability and reliability of the precision load control system, the Precision load control system needs to be fully verified before it is put into use. In this paper, a set of simulation test environment is developed for the characteristics of precise load control system devices and complex test systems. The test environment is based on the actual software and hardware architecture, for internal communication characteristics, research test methods, customized test case drivers, and simulation. Various operating conditions under full load operation, comprehensive verification and assessment of the functional performance of the precision load control system to ensure the stability and reliability of the precision load control system. The proposed simulation test method has certain reference significance for the simulation test of precise load control system and new energy safety and stability control system.

This paper introduces the design of a precise load shedding system and its simulation test method. The precision load shedding system is part of the grid safety and stability control system. It is mainly used to coordinate the comprehensive configuration of power grid resources and loads, realize mutual interaction between power supply, power grid and user load, and improve the emergency response capability of the grid for UHV faults. Through the classification, grading and sub-regional management of the load resources, the real-time precise control of the interruptible load of the classified users can be realized by the dispatching direct command. This paper mainly studies the architecture, strategy and communication protocols of each level of the system. In addition, the algorithm of the load shedding strategy of the control terminal was studied. Finally, a simulation test method for precise load shedding system is proposed in RTDS environment, which has certain reference significance for the safety and stability control of large power grid, micro-grid and distributed energy.

System control and the collection of analog signals are fundamental tasks in many embedded computer systems such as are found in automotive, communication, and sensor network domains. Often latency is critical and FPGAs are an attractive alternative. Traditionally, external ADC (analog to digital converter) chips have been used for analog to digital signal conversion and transfer of the digital signals to the FPGA(s). In large-scale quantum system experiments, the implementation of this classic control infrastructure is a challenge. In particular, the FPGA-based control system must work in a liquid helium environment at about 4°K. To improve the system's reliability, we need to integrate multiple lanes of soft core ADC directly into the FPGA. In this paper we propose a method of building high-speed ADCs with time to digital converters (TDCs). The experimental results show that the ADCs can achieve a sampling rate of 100Msa/s with a 6-bit resolution for signals ranging from 0 to 3 V. In our design the ADC uses primarily the ISERDES logic of a Xilinx FPGA plus a small number of CLBs. Our design can integrate 24 lanes of soft-core ADCs into a Xilinx XC7A100t-2csg324 FPGA.

Min-product fuzzy relation inequalities is introduced to characterise the pricing relation in a supply chain system. In such system, all the retailers are supplied with some kind of target commodities. Meanwhile, every supplers should be able to supply its local commodities to at least one retailer. For the min-product fuzzy relation system, we focus on the relevant properties which are helpful for the resolution. Specific structure and characteristic of the complete solution set are also investigated. Numerical examples are presented to illustrate the system consistency.

Background In the post-genome age, it is more urgent to understand the functions of genes and proteins. Since experimental methods are usually costly and time consuming, computational predictions are recognized as an alternative approach. In developing a predictive method for functional genomics and proteomics, one of the most important steps is to represent biological sequences with a fixed length numerical form, which can be further analyzed using machine learning algorithms. Chou's pseudo-amino acid compositions and the pseudo k-nucleotide compositions are algorithms for this purpose. Conclusion Since the appearance of these algorithms, several software tools have been developed as implementations. These software tools facilitate the application of these algorithms. As these software tools are developed with different technologies and for different application scenarios, we will briefly review the technical aspect of these software tools in this short review.

In the field of human-computer interaction, automatic emotion recognition is an important and challenging task. As a physiological signal that directly reflects the brain activity, EEG has advantages in emotion recognition. However, previous studies seldom consider together the temporal, spatial, and frequency characteristics of EEG signals, and the reported emotion recognition accuracy is not adequate for applications. To address this issue, this study proposes a new approach which extracts RASM as the feature to describe the frequency-space domain characteristics of EEG signals and constructs a LSTM network as the classifier to explore the temporal correlations of EEG signals. It is implemented on the DEAP dataset for a trial-level emotion recognition task. In a comparison with a number of relevant studies on DEAP, its mean accuracy of 76.67% ranks the first, which approves the effectiveness of this new approach.

A novel computer-aided method based on magnetic resonance images (MRI) was proposed for the early detection and diagnosis of nasopharyngeal carcinoma (NPC). A local Chan-Vese level-set model, which integrated the maximum interclass-variance method with the Chan-Vese model, was built to detect foci with unobvious boundaries. For each of the suspected foci, 26 features, including suspected focus texture, shape, and grayscale characteristics, were extracted, and then classified with a support-vector-machine (SVM) classifier. The method was tested with 289 brain images of 48 patients with nasopharyngeal carcinoma and 33 healthy adults, which obtained an average successful-diagnosis rate of 90.74%.

In this paper, we present a new method to categorize the heart sound signals into Normal, Murmur, Extra Heart Sound and Artifact with Convolutional Neural Network (CNN) model. The experimental dataset are from the PASCAL heart sound classification workshop, 2011. Our method mainly contains three steps as follows. Firstly, we propose cross- cutting and framing to enlarge the data. The data used for training and test is three times than before. Secondly, we manage the heart sound signals via Butterworth filter and Fast Fourier Transform simply. Finally the captured frequency features are input into the CNN model for classification. Compared with the traditional heart sounds classification approaches-selecting features and designing classifier, our method overcomes the uncertain in selecting features and improves the self-learning ability. The experimental results show that deep learning network model have higher classification effectiveness for the data with much noise. The global identification rate reaches 98%, more effective than previous related studies.

This paper presents a vision-assisted human motion analysis method by which to classify unknown preparatory motions for bed exit. A convolutional neural network was applied to identify the existence of essential postures between two meaningful sequential motions from low-resolution depth images. The preliminary experimental results showed that sequential images of bed exit preparatory motions can be classified as essential motion states, which is useful for construction of the states of a hidden Markov model for bed exit prediction.

This paper studies a hybrid artificial bee colony (ABC) algorithm for finding high quality solutions of the job-shop scheduling problem with no-wait constraint (JSPNW) with the objective of minimizing makespan among all the jobs. JSPNW is an extension of well-known job-shop scheduling problem subject to the constraint that no waiting time is allowed between operations for a given job. ABC algorithm is a swarm intelligence technique based on intelligent foraging behavior of honey bee swarm. The proposed hybrid approach effectively coordinates the various components of ABC algorithm such as solution initialization, selection and determination of a neighboring solution with the local search in such a way that it leads to high quality solutions for the JSPNW. The proposed approach is compared with the two best approaches in the literature on a set of benchmark instances. Computational results show the superiority of the proposed approach over these two best approaches.

A quantum realization of the generalized Arnold transform is designed. A novel quantum image encryption algorithm based on generalized Arnold transform and double random-phase encoding is proposed. The pixels are scrambled by the generalized Arnold transform, and the gray-level information of images is encoded by the double random-phase operations. The keys of the encryption algorithm include the independent parameters of coefficients matrix, iterative times and classical binary sequences, and thus, the key space is extremely large. Numerical simulations and theoretical analyses demonstrate that the proposed algorithm with good feasibility and effectiveness has lower computational complexity than its classical counterpart.

Background Typical radio frequency identification (RFID) access control system can be ineffective if an unauthorized person tailgates an authorized person through an access area. Objective To propose a system by using indoor locating and tracking techniques address this problem, which is to prevent unauthorized Alzheimer's and dementia patients from getting lost including by tailgating. Methods To achieve accurate target location, passive RFID deployment strategy is studied and a fingerprinting based passive RFID localization algorithm is proposed. The proposed system was evaluated in a building environment to simulate the performance of access control. RFID reader was installed on ceiling near the access area and tags were stitched on both shoulders of the experiment subject's garments. Results The probability of the error distance within 0.3 m achieved 97% in the warning area; the location precision achieved 97% within 0.4 m in the monitoring area. Conclusions The result showed that if an unauthorized person enters the restricted area, the system can initiate an alert signal accurately. Therefore, the proposed system is very suitable to be used in nursing home or hospital to prevent unauthorized personnel and assets entering/exiting a confined location.

In this paper, we propose a deep convolutional neural network model for in-bed behavior recognition and bed-exit prediction. This model extracts features for training from depth images taken by depth cameras in two categories: in-bed images taken several time intervals before a patient gets out of bed, and usual in-bed activity images. The depth camera-based model features grayscale and low-resolution images, and excels at discarding unnecessary details such as background information beyond the target's contour. The proposed model was proven to be computationally efficient, which is crucial for analyzing and predicting the category of new images in real time.

This paper selects bias slider-crank mechanism with a constant angular velocity as the research object. First the motion law of mechanism is derived by adopting complex vector method, and then the motion law curve of slider is solved and drawn with MATLAB which can accurately reflect the actual slider motion law; meanwhile output ADAMS simulation result data and compare ADAMS simulation results with the derived results with MATLAB. The results show slider motion law of ADAMS simulation coincides with the theoretical derivation result, which is simulation intuitive and easy to measure.

Cloud computing is the latest trend in IT technical development, the importance of cloud databases has been widely acknowledged. There are numerous data in the cloud database and among these data, much potential and valuable knowledge are implicit. The key point is to discover and pick up the useful knowledge automatically. An association rule is one of the main models in mining out these data, and it mainly focuses on the relationship among different areas in the data. This paper puts forward the basic model of data mining based on association rules in cloud database and introduces corresponding mining algorithms.

Most of Frame Buffer Compression (FBC) methods are based on variable-length compression algorithm. The proposed fixed-ratio frame buffer compressor, which is based on S3 texture compression (S3TC) algorithm, compresses the recently rendered image into DXT format so that it can be employed as input compressed texture. We have optimized the frame buffer compressor to achieve real time performance and FPGA implementation, which occupies 406 slice registers and 1133 slice LUT in Xilinx Spartan6 FPGA. The compressor consumes 10% additional silicon space, but it reduces the bandwidth consumption to 16.8%–45.8% compared to that of 50–90% by the variable-length compression methods.

This paper presents constraint generalized orthogonal matching pursuit (C-gOMP) approach to tackle the problem of clutter mitigation for radar imaging in the compressive sensing context. The generalized orthogonal matching pursuit algorithm (gOMP) in radar imaging process is susceptible to environmental noise interference. Therefore, a particular constraint condition is added to the recovery signal, and the noise component is restrained after the greedy iteration. The C-gOMP algorithm uses the cost function to impose more constraints on each coefficient to ensure the convergence of the whole function. To further analyze the performance of C-gOMP algorithm, the effectiveness of this algorithm is demonstrated by simulations. The results show that the proposed approach is very obvious at suppressing unwanted clutter and enhancing the desired targets, and C-gOMP algorithm brings an evident performance improvement and application prospect.

In this paper, we propose a new optical flow-based Poisson inverse gradient (OFPIG) initialization method for active contour tracking. This method can automatically initialize the contour of moving target for consequent tracking. First, an optical flow based motion detection method is adopted to remove background information, and then a Poisson inverse gradient (PIG) initialization is applied to locate the target region. Finally, parametric active contour is used to evolve the correct target contour depending on the initialization precision. Experimental results have demonstrated its effectiveness and robustness.

According to Assemble line balancing (ALB) problem which is widespread in manufacture, we establish a mathematical model and present our improved algorithm that is based on Heuristic rules, which adopt cumulative ranked positional weight as a basis of task assignments and schedule the orders of branches traversal. After that we can find an optimal solution to the balancing problem efficiently. The result of examples given by this paper shows our algorithm has a good performance.

There is an immense amount of data in the cloud database and among these data, much potential and valuable knowledge are implicit. The key point is to discover and pick out the useful knowledge, and to do so automatically. In this paper, the data model of the cloud database is analyzed. The relationships among different areas in the data are then analyzed, from which the new knowledge can be found. The basic data mining model based on the cloud database is defined, and the discovery algorithm is presented.

Based on Hilbert transform and complex trigonometric operation, this paper presents a new digital algorithm to improve the monitoring accuracy of the dielectric loss factor, and the theoretical basis of the algorithm is also deduced. Simulation results show that, the new algorithm renders good application performance when there exists power frequency fluctuation or the sampled data length is not exactly an integer fold of a power cycle, which demonstrates the algorithm can effectively solve the issues incurred from asynchronous sampling.