Search

Video monitoring is an important means to ensure production safety in coal mine. However, the currently intelligent video surveillance is difficult to respond in real-time due to the latency of cloud computing. In this paper, a cloud-edge cooperation framework is proposed, which integrates cloud computing and edge computing in a coordinated manner. The cloud computing is used to process non-real-time and global tasks, while the edge computing is responsible for handling local monitoring video in real-time. In order to realize cloud-edge data interaction and online optimization for edge models, the heterogeneous converged network is built. In addition, an object detection model FL-YOLO composed of depthwise separable convolution and down-sampling inverted residual block is proposed, which realizes real-time video analysis at the edge. Finally, this paper discusses the complexity of FL-YOLO by its computational cost and model size. The experiment results show that the model size of FL-YOLO is 16.1MB, which is very light, and it achieves 36.7 FPS on NVIDIA Jetson TX1 and an AP of 76.7% on Multi-scene pedestrian dataset. Comparing with mainstream object detection models, FL-YOLO completes faster detection speed and higher accuracy, and it has lower calculation complexity and smaller model scale. Furthermore, the AP on Single-scene pedestrian dataset of FL-YOLO is improved to 80.7% by cloud-edge cooperation. K-Fold method is also used to further compared the performance of FL-YOLO and other models. Moreover, system test is implemented on coal mine, which validates the actual engineering effect of the proposed cloud-edge cooperation framework.

In this article, we design and implement a full-packet capture and network measurement (FPC-NM) system in a 20 Gbit/s network. The system has multiple functions, such as packet receiving, nanosecond timestamp, load balancing, data packet preprocessing, application layer protocol analysis, data packet storage, and log management. We design a timestamp module in field programmable gate array (FPGA) that can add a timestamp to each captured packet, and the retrospective packet analysis can be done with 4 ns accuracy. As far as we know, this is the first time that nanosecond timestamp has been applied to FPC-NM system, which greatly improves the accuracy of security incident retrospective analysis. The experimental results show that the FPC-NM system can achieve 17 Gbit/s throughput and 160 000 connections per second under the premise of zero packet loss. Using LZ4 compression, the performance of our system for real-time compression and storage of traffic can achieve 10 Gbit/s. The system performance is improved by about 40% after software optimization including direct memory access buffer size adjustment and single instruction, multiple data (SIMD) instruction set, which validates the practicability and effectiveness of our optimization method. [ABSTRACT FROM AUTHOR]

Steady-state analysis on production systems, especially serial lines with rigid production mode, have been carried out in a large amount of research literature. With the development of the advanced manufacturing technologies, as well as the national strategies, for example, Made in China 2025, real-time performance evaluation, prediction, production management, and efficient control of systems with flexible production mode are gradually becoming important issues. Under the assumption of assembly systems with small batch-based production tasks, we develop a mathematical model for the real-time performance analysis of such systems with the geometric machine reliability model, which is applicable in many manufacturing environments. The formulas for real-time performance analysis are derived first using the Markovian approach for three-machine assembly systems. Then, based on the idea of decomposition and aggregation, an algorithm with high accuracy and efficiency for system real-time performance approximation is proposed. Finally, the decomposition and aggregation-based real-time analysis approach is extended to the generalized assembly systems. Note to Practitioners—As the concept of manufacturing intelligence being proposed and promoted, system flexibility appears frequently in production in recent years. A flexible production refers to the implementation of small batch-based production with multitype of products on the same production line, by quick setup or machine adjustments. The steady-state analysis method that is traditionally used might be not applicable in this case since most of the production process is in transient state. An approximation-based analytical method, for real-time performance evaluation of assembly systems with small batch-based production tasks and with machines having the geometric reliability model, is proposed. The algorithm developed in this article can be utilized by production managers and engineers, with which the real-time performance of the systems can be predicted with high accuracy. It can also provide assistance for decision-making on activities of production control. [ABSTRACT FROM AUTHOR]

In this article, self-adaptive variational mode decomposition (SAVMD), deep convolutional neural networks (DCNN), and online-sequential random vector functional link networks (OSRVFLN) are integrated to categorize the single as well as combined power quality events (PQEs) in real time. The SAVMD method is proposed to optimize both the number of decomposition and data-fidelity factor to extract the most efficient band-limited mode (BLM) based on entropy and Kurtosis index. The most discriminative unsupervised features are extracted automatically using a DCNN from the BLM of SAVMD. The extracted distinct feature vector is fed to the proposed supervised OSRVFLN classifier to train accurately by minimizing the training cross-entropy loss with an increment in the number of hidden nodes for obtaining the maximum classification accuracy of the complex PQE patterns in noisy and noise-free environments. The automatic efficacious feature extraction, superior classification accuracy, noise immunity, and short event detection time are the major advantages of the proposed SAVMD-DCNN-OSRVFLN method. Finally, the novel methodology is implemented in a fast digital Xilinx Virtex-5 field-programmable gate array embedded processor to validate the practicability and feasibility of the proposed method in real-time. [ABSTRACT FROM AUTHOR]

An extensive collection of algorithms have been proposed over the years to identify the foreground from a video scene, but none of them considers the classification history of previous frames for discovering moving objects. All the existing algorithms focus on a single background model to cope with all types of challenging and complex video environments. In this paper, a real-time pixel level classification method is proposed that uses its previous output history to update its parameters. The model has three components designed to handle various types of challenging background environments. For each pixel, three sub-parts, namely the Adaptive Background Model, the Neighborhood Background Model, and the Change Detection Background Model are constructed to detect various types of complex background changes. A pixel level model updating method uses the previous foreground/background binary classification results to periodically refresh all the three background models. This update method helps to recognize the foreground accurately by adjusting the algorithmic parameters to efficiently detect complex background changes. This novel method shows a significant improvement in performance for a variety of complex video scenes. [ABSTRACT FROM AUTHOR]

Our goal was to develop an automated system to determine whether animals have learned and changed their behavior in real-time using a low calculation-power central processing unit (CPU). The bottleneck of real-time analysis is the speed of image recognition. For fast image recognition, 99.5% of the image was excluded from image recognition by distinguishing between the subject and the background. We achieved this by applying a binarization and connected-component labeling technique. This task is important for developing a fully automated learning apparatus. The use of such an automated system can improve the efficiency and accuracy of biological studies. The pond snail Lymnaea stagnails can be classically conditioned to avoid food that naturally elicits feeding behavior, and to consolidate this aversion into long-term memory. Determining memory status in the snail requires real-time analysis of the number of bites the snail makes in response to food presentation. The main algorithm for counting bites comprises two parts: extracting the mouth images from the recorded video and measuring the bite rate corresponding to the memory status. Reinforcement-supervised learning and image recognition were used to extract the mouth images. A change in the size of the mouth area was used as the cue for counting the number of bites. The accuracy of the final judgment of whether or not the snail had learned was the same as that determined by human observation. This method to improve the processing speed of image recognition has the potential for broad application beyond biological fields. [ABSTRACT FROM AUTHOR]

In automotive embedded real-time systems, e.g., the engine control unit, tasks are activated whenever the engine arrives at a specific angular position. In consequence the frequency of activation changes with the speed of the engine. Additionally, the execution times and deadlines may also depend on the engine speed. In recently published works analytical methods for tasks with this rate-dependent behavior were introduced. Though those methods do not consider angular phases between tasks. For instance, in a typical engine control unit one event is displaced a certain angular position after an event of another task. This paper introduces a sufficient analysis, which considers such angular phases. A number of experimental results are reported to compare our method with existing analysis. These illustrate the improvement in terms of precision and runtime. [ABSTRACT FROM AUTHOR]

In this paper, Hilbert Huang transform (HHT) and weighted bidirectional extreme learning machine (WBELM) are integrated to detect and classify power quality events (PQEs) in real time. Empirical mode decomposition is used to decompose the nonstationary PQEs into the monocomponent mode of oscillation, known as intrinsic mode functions (IMFs). The efficacious features are extracted by applying the Hilbert transform (HT) on the IMFs. An efficient WBELM computational intelligence technique is proposed to recognize the single, as well as multiple PQEs and its performances are compared with the recently developed classifiers such as support vector machine, least-square support vector machine, extreme learning machine, and bidirectional extreme learning machine. The recognition architecture of HHT integrated with WBELM (HHT-WBELM) method is tested and compared with the empirical wavelet transform associated with HT and WBELM method, and tunable-Q wavelet transform along with HT and WBELM method. The faster learning speed, lesser computational complexity, superior classification accuracy, and short event detection time prove that the proposed HHT-WBELM method can be implemented in the online power quality monitoring system. Finally, a hardware prototype is developed based on the digital signal processor to verify the cogency of the proposed method in real time. The feasibility of the proposed method is tested and validated by both the simulation and laboratory experiments. [ABSTRACT FROM AUTHOR]

This paper proposes an efficient real-time background subtraction algorithm, which is essential in many computer vision applications. Initially, histograms of the intensity values for each channel of a pixel position in a set of training framesare constructed. A background model, histogram min-max bucket, is constructed from the minimum and maximum values of contiguous non-zero frequencies of the temporal intensity histogram. A novel feature of this algorithm is the use of a single sliding window to update the system adaptively, capturing the proper background even under sudden and/or gradual illumination changes in the scene. We incorporate both local and global information with the help of the current and the previously visited pixel values for binary classification of a pixel into foreground and background. This algorithm is compared with several state-of-the-art techniques and experimental studies show that the proposed method outperforms all these methods in terms of accurate binary classification. [ABSTRACT FROM AUTHOR]

Pulse wave analysis is an important method used to gather information about the cardiovascular system. Instead of detecting the pulse wave via pressure sensors, bioimpedance measurements can be performed to acquire minuscule changes in the conductivity of the tissue, caused by the pulse wave. This work presents a microcontroller-based bioimpedance measurement system, which has the capability to acquire impedance measurements from up to four independent channels simultaneously. By combining a problem-specific analog measurement circuit with a 24 bits analog-to-digital converter, the system is capable of acquiring 1000 impedances per second with a signal-to-noise ratio in a range from 92 to 96 dB. For data storage and analysis, the digitized data are sent via universal serial bus to a host PC. A graphical user interface filters and plots the data of all channels in real-time. The performance of the system regarding measuring constant impedances, as well as impedance changes over time is demonstrated. Two different applications for pulse wave analysis via multichannel bioimpedance measurements are presented. Additionally, first measurement results from a human subject are shown to demonstrate the system's applicability of analyzing the pulse wave morphology as well as the aortic pulse wave velocity. [ABSTRACT FROM AUTHOR]

In embedded real-time systems, the schedulability analysis is an important method to verify whether the real-time constraints are satisfied. Especially, engine control systems in the automotive industry are particular challenging regarding their real-time analysis. Some of the tasks of such systems are triggered at predetermined angular values of the crankshaft. Thus, the frequency of activation increases for higher crankshaft rotation speeds. To avoid overloads, these tasks reduce their computational demand for higher rotation speeds. The variable rate-dependent behavior (VRB) task model has been proposed as a means of modeling such rate-dependent behavior. In this paper, we introduce a new method for the exact computation of interference for VRB-tasks. A number of experimental results are reported to validate our approach and compare it with existing analysis. These experimental results illustrate that our method outperforms previous results in terms of precision and runtime. [ABSTRACT FROM PUBLISHER]

The amount of electronic systems in agriculture machinery grows, becoming more complex and multifunctional. Due to the increasing number of devices connected to the network and the complexity of their functions, the current controller area network (CAN) becomes overloaded and CAN with flexible data rate (CAN FD) emerges as an alternative. This paper presents the simulation of an agriculture machine vehicular network using the higher layer protocols J1939 and ISO11783 (ISOBUS) and the performance comparison of the network using CAN, CAN FD 8 B and CAN FD 64 B. The presented contribution is mainly twofold: the proposal of an SAE J1939/ISOBUS compliant CAN FD 64 B model for a highly demanding vehicular network, and a detailed analysis of the real-time properties of this model compared to CAN and CAN FD 8 B. In the performed experiments, on the network permanent regime, periodic messages of machine control functions and network management data were sent by the devices connected to the network. The bus load usage, the response time, and jitter of these messages were analyzed to evaluate the real-time characteristics of the network. Compared to CAN, CAN FD had faster object pool transference, lower bus load usage, shorter worst case response time, and lower jitter. [ABSTRACT FROM PUBLISHER]

In this work, a WAMS-Based Voltage Stability Indicator considering real time operation conditions is proposed. The indicator considers operational limits, such as reactive power reserves, transmission constraints, generating capacity, and voltage limits. The proposed indicator is based on a combination of the reactive margin and the actual demand at each busbar of the system. The reactive margins are obtained from an optimization problem. Applications on a test system show that the indicator is able to identify the proximity to voltage collapse, successfully. It also provides a reliable measure to assess the system security of real time operation in power grids. [ABSTRACT FROM PUBLISHER]

This paper proposes a real-time power quality disturbances (PQDs) classification by using a hybrid method (HM) based on S-transform (ST) and dynamics (Dyn). Classification accuracy and run time are mainly considered in our work. The HM firstly uses Dyn to identify the location of the signal components in the frequency spectrum yielded by Fourier transform, and uses inverse Fourier transform to only some of the signal components. Then features from Fourier transform, ST, and Dyn are selected, and a decision tree is used to classify the types of PQD. In order to reduce the influence of Heisenberg' s uncertainty, we proposed that different signal components are windowed by different Gaussian windows, which brings better adaption and flexibility. By the HM, run time of the application has been greatly reduced with satisfactory classification accuracy. Finally, a DSP-FPGA based hardware platform is adopted to test the run time and correctness of the proposed method under real standard signals. Field signal tests have also presented. Both simulations and experiments validate the feasibility of the new method. [ABSTRACT FROM AUTHOR]

This paper presents the design and development of a hardware-based laboratory smart grid test-bed. This system is developed at the Energy Systems Research Laboratory, Florida International University. The hardware/software based system includes implementation of control strategies for generating stations, and power transfer to programmable loads in a laboratory scale of up to 35 kilowatts in ac power and 36 kW in renewable sources and energy storages. Appropriate software was developed to monitor all system parameters as well as operate and control the various interconnected components in varying connectivity architectures. The interconnection of alternate energy such as wind emulators, PV arrays, and fuel cell emulators are implemented, studied and integrated into this system. Educational experiences were drawn during the design and system development of this laboratory-based smart grid. The real-time operation and analysis capability provides a platform for investigation of many challenging aspects of a real smart power system. The design, development, and hardware setup of this laboratory is presented here in Part I of this paper. This includes component development, hardware implementation, and control and communication capabilities. Part II of the paper presents the implementation of the monitoring, control, and protection system of the whole setup with detailed experimental and simulation results. [ABSTRACT FROM PUBLISHER]

Sensors of various modalities and capabilities, especially cameras, have become ubiquitous in our environment. Their intended use is wide ranging and encompasses surveillance, transportation, entertainment, education, healthcare, emergency response, disaster recovery, and the like. Technological advances and the low cost of such sensors enable deployment of large-scale camera networks in large metropolises such as London and New York. Multimedia algorithms for analyzing and drawing inferences from video and audio have also matured tremendously in recent times. Despite all these advances, large-scale reliable systems for media-rich sensor-based applications, often classified as situation-awareness applications, are yet to become commonplace. Why is that? There are several forces at work here. First, the system abstractions are just not at the right level for quickly prototyping such applications on a large scale. Second, while Moore's law has held true for predicting the growth of processing power, the volume of data that applications are called upon to handle is growing similarly, if not faster. Enormous amount of sensing data is continually generated for real-time analysis in such applications. Further, due to the very nature of the application domain, there are dynamic and demanding resource requirements for such analyses. The lack of right set of abstractions for programing such applications coupled with their data-intensive nature have hitherto made realizing reliable large-scale situation-awareness applications difficult. Incidentally, situation awareness is a very popular but ill-defined research area that has attracted researchers from many different fields. In this paper, we adopt a strong systems perspective and consider the components that are essential in realizing a fully functional situation-awareness system. [ABSTRACT FROM AUTHOR]

A novel increased group-delay slope loop scheme is proposed to enhance the time-frequency resolution of dispersive delay structure (DDS) components for microwave analog signal processing systems. In this scheme, the signal at the output of the DDS is regenerated by an amplifier and reinjected via a nondispersive delay line toward its input along a loop. At each pass across the DDS, the effective group-delay slope of the system is increased so that, after n turns along the loop, the time-frequency resolution has been enhanced by a factor n. This approach provides a solution to the unpractical approach of cascading n DDS units, which would lead to excessively large device footprint, unacceptably high insertion loss, and severe signal-to-noise reduction. The proposed scheme is implemented in a proof-of-concept circuit using a C-section all-pass network DDS and demonstrated experimentally in a frequency meter and in a frequency discriminator. Possible improvements for higher performance are discussed. [ABSTRACT FROM PUBLISHER]

Heart rate variability (HRV) is traditionally analyzed while a subject is in a controlled environment, such as at rest in a clinic, where it can be used as a medical indicator. This paper concerns analyzing HRV outside of controlled environments, such as on an actively moving person. We describe automated methods for inter-heartbeat interval (IBI) error detection and correction. We collected 124,998 IBIs from 18 subjects, undergoing a variety of active motions, for use in evaluating our methods. Two human graders manually labeled each IBI, evaluating 10% of the IBIs as having an error, which is a far greater error percentage than has been examined in any previous study. Our automated method had a 96% agreement rate with the two human graders when they themselves agreed, with a 49% rate of matching specific error corrections and a 0.01% false alarm rate. [ABSTRACT FROM AUTHOR]

Time-wavelength spectroscopy, using a linearly chirped broad-band pulse, eliminates the need for an optical spectrometer and enables real-time analysis. Utilizing digital filtering and matched detection, we demonstrate identification of gas signatures that are entirely masked by highly nonuniform source spectra. [ABSTRACT FROM PUBLISHER]

This article describes the development of adaptive testing in response to the ever-growing need to dynamically and cost-effectively tailor IC testing to discriminately manage manufacturing process variations. Various degrees of adoption are presented, together with benefits and examples of its use. Finally, challenges for future development are discussed. [ABSTRACT FROM PUBLISHER]