Binary neural network based real time emotion detection on an edge computing device to detect passenger anomaly

With the increase in research cases of the application of a convolutional neural network (CNN)-based object detection technology, studies on the light-weight CNN models that can be performed in real time on the edge-computing devices are also increasing. This paper proposed scalable convolutional blocks that can be easily designed CNN networks of You Only Look Once (YOLO) detector which have the balanced processing speed and accuracy of the target edge-computing devices considering different performances by exchanging the proposed blocks simply. The maximum number of kernels of the convolutional layer was determined through simple but intuitive speed comparison tests for three edge-computing devices to be considered. The scalable convolutional blocks were designed in consideration of the limited maximum number of kernels to detect objects in real time on these edge-computing devices. Three scalable and fast YOLO detectors (SF-YOLO) which designed using the proposed scalable convolutional blocks compared the processing speed and accuracy with several conventional light-weight YOLO detectors on the edge-computing devices. When compared with YOLOv3-tiny, SF-YOLO was seen to be 2 times faster than the previous processing speed but with the same accuracy as YOLOv3-tiny, and also, a 48% improved processing speed than the YOLOv3-tiny-PRN which is the processing speed improvement model. Also, even in the large SF-YOLO model that focuses on the accuracy performance, it achieved a 10% faster processing speed with better accuracy of 40.4% [email protected] in the MS COCO dataset than YOLOv4-tiny model.

Download Full-text

A Multifault Diagnosis Method of Gear Box Running on Edge Equipment

Security and Communication Networks ◽

10.1155/2020/8854236 ◽

2020 ◽

Vol 2020 ◽

pp. 1-13

Author(s):

Xiao-ping Zhao ◽

Yong-hong Zhang ◽

Fan Shao

Keyword(s):

Fault Diagnosis ◽

Real Time ◽

Data Center ◽

Edge Computing ◽

Industrial Equipment ◽

Point Model ◽

Computing Device ◽

External Data ◽

Diagnosis Method ◽

Analyze Data

In recent years, a large number of edge computing devices have been used to monitor the operating state of industrial equipment and perform fault diagnosis analysis. Therefore, the fault diagnosis algorithm in the edge computing device is particularly important. With the increase in the number of device detection points and the sampling frequency, mechanical health monitoring has entered the era of big data. Edge computing can process and analyze data in real time or faster, making data processing closer to the source, rather than the external data center or cloud, which can shorten the delay time. After using 8 bits and 16 bits to quantify the deep measurement learning model, there is no obvious loss of accuracy compared with the original floating-point model, which shows that the model can be deployed and reasoned on the edge device, while ensuring real time. Compared with using servers for deployment, using edge devices not only reduces costs but also makes deployment more flexible.

Download Full-text

Edge Computing Based IoT Architecture for Low Cost Air Pollution Monitoring Systems: A Comprehensive System Analysis, Design Considerations & Development

Sensors ◽

10.3390/s18093021 ◽

2018 ◽

Vol 18 (9) ◽

pp. 3021 ◽

Cited By ~ 26

Author(s):

Zeba Idrees ◽

Zhuo Zou ◽

Lirong Zheng

Keyword(s):

Air Quality ◽

Power Consumption ◽

Real Time ◽

Low Cost ◽

Edge Computing ◽

Quality Monitoring ◽

Quality Data ◽

Monitoring Systems ◽

Air Quality Monitoring ◽

Computing Device

With the swift growth in commerce and transportation in the modern civilization, much attention has been paid to air quality monitoring, however existing monitoring systems are unable to provide sufficient spatial and temporal resolutions of the data with cost efficient and real time solutions. In this paper we have investigated the issues, infrastructure, computational complexity, and procedures of designing and implementing real-time air quality monitoring systems. To daze the defects of the existing monitoring systems and to decrease the overall cost, this paper devised a novel approach to implement the air quality monitoring system, employing the edge-computing based Internet-of-Things (IoT). In the proposed method, sensors gather the air quality data in real time and transmit it to the edge computing device that performs necessary processing and analysis. The complete infrastructure & prototype for evaluation is developed over the Arduino board and IBM Watson IoT platform. Our model is structured in such a way that it reduces the computational burden over sensing nodes (reduced to 70%) that is battery powered and balanced it with edge computing device that has its local data base and can be powered up directly as it is deployed indoor. Algorithms were employed to avoid temporary errors in low cost sensor, and to manage cross sensitivity problems. Automatic calibration is set up to ensure the accuracy of the sensors reporting, hence achieving data accuracy around 75–80% under different circumstances. In addition, a data transmission strategy is applied to minimize the redundant network traffic and power consumption. Our model acquires a power consumption reduction up to 23% with a significant low cost. Experimental evaluations were performed under different scenarios to validate the system’s effectiveness.

Download Full-text

Short-Term Traffic State Prediction Based on Mobile Edge Computing in V2X Communication

Applied Sciences ◽

10.3390/app112311530 ◽

2021 ◽

Vol 11 (23) ◽

pp. 11530

Author(s):

Pangwei Wang ◽

Xiao Liu ◽

Yunfeng Wang ◽

Tianren Wang ◽

Juan Zhang

Keyword(s):

Neural Network ◽

Real Time ◽

Edge Computing ◽

Mobile Edge Computing ◽

Short Term ◽

Spatiotemporal Features ◽

Traffic State ◽

State Prediction ◽

Comparison Results ◽

Perception System

Real-time and reliable short-term traffic state prediction is one of the most critical technologies in intelligent transportation systems (ITS). However, the traffic state is generally perceived by single sensor in existing studies, which is difficult to satisfy the requirement of real-time prediction in complex traffic networks. In this paper, a short-term traffic prediction model based on complex neural network is proposed under the environment of vehicle-to-everything (V2X) communication systems. Firstly, a traffic perception system of multi-source sensors based on V2X communication is proposed and designed. A mobile edge computing (MEC)-assisted architecture is then introduced in a V2X network to facilitate perceptual and computational abilities of the system. Moreover, the graph convolutional network (GCN), the gated recurrent unit (GRU), and the soft-attention mechanism are combined to extract spatiotemporal features of traffic state and integrate them for future prediction. Finally, an intelligent roadside test platform is demonstrated for perception and computation of real-time traffic state. The comparison experiments show that the proposed method can significantly improve the prediction accuracy by comparing with the existing neural network models, which consider one of the spatiotemporal features. In particular, for comparison results of the traffic state prediction and the error value of root mean squared error (RMSE) is reduced by 39.53%, which is the greatest reduction in error occurrences by comparing with the GCN and GRU models in 5, 10, 15 and 30 minutes respectively.

Download Full-text

Real Time Face Recognition on an Edge Computing Device

Proceedings of the 2020 9th International Conference on Software and Computer Applications ◽

10.1145/3384544.3384567 ◽

2020 ◽

Author(s):

Samarth Gupta

Keyword(s):

Face Recognition ◽

Real Time ◽

Edge Computing ◽

Computing Device

Download Full-text

Edge computing-based real-time passenger counting using a compact convolutional neural network

Neural Computing and Applications ◽

10.1007/s00521-018-3894-2 ◽

2018 ◽

Vol 32 (9) ◽

pp. 4919-4931 ◽

Cited By ~ 1

Author(s):

Biao Yang ◽

Jinmeng Cao ◽

Xiaofeng Liu ◽

Nan Wang ◽

Jidong Lv

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Real Time ◽

Edge Computing

Download Full-text

Driver fatigue detection based on convolutional neural network and face alignment for edge computing device

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1177/0954407021999485 ◽

2021 ◽

pp. 095440702199948

Author(s):

Xiaofeng Li ◽

Jiahao Xia ◽

Libo Cao ◽

Guanjun Zhang ◽

Xiexing Feng

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Face Detection ◽

Edge Computing ◽

Face Alignment ◽

Driver Fatigue ◽

Computing Device ◽

The Face ◽

Yaw Angle ◽

Fatigue Detection

Most current vision-based fatigue detection methods don’t have high-performance and robust face detector. They detect driver fatigue using single detection feature and cannot achieve real-time efficiency on edge computing devices. Aimed at solving these problems, this paper proposes a driver fatigue detection system based on convolutional neural network that can run in real-time on edge computing devices. The system firstly uses the proposed face detection network LittleFace to locate the face and classify the face into two states: small yaw angle state “normal” and large yaw angle state “distract.” Secondly, the speed-optimized SDM algorithm is conducted only in the face region of the “normal” state to deal with the problem that the face alignment accuracy decreases at large angle profile, and the “distract” state is used to detect driver distraction. Finally, feature parameters EAR, MAR and head pitch angle are calculated from the obtained landmarks and used to detect driver fatigue respectively. Comprehensive experiments are conducted to evaluate the proposed system and the results show its practicality and superiority. Our face detection network LittleFace can achieve 88.53% mAP on AFLW test set at 58 FPS on the edge computing device Nvidia Jetson Nano. Evaluation results on YawDD, 300 W, and DriverEyes show the average detection accuracy of the proposed system can reach 89.55%.

Download Full-text

Physical Distancing Device with Edge Computing for COVID-19 (PADDIE-C19)

Sensors ◽

10.3390/s22010279 ◽

2021 ◽

Vol 22 (1) ◽

pp. 279

Author(s):

Chun Hoe Loke ◽

Mohammed Sani Adam ◽

Rosdiadee Nordin ◽

Nor Fadzilah Abdullah ◽

Asma Abu-Samah

Keyword(s):

Neural Network ◽

Close Contact ◽

Low Cost ◽

Absolute Error ◽

Face Mask ◽

Physical Distance ◽

Edge Computing ◽

Computing Device ◽

Person Detection ◽

The Neural Network

The most effective methods of preventing COVID-19 infection include maintaining physical distancing and wearing a face mask while in close contact with people in public places. However, densely populated areas have a greater incidence of COVID-19 dissemination, which is caused by people who do not comply with standard operating procedures (SOPs). This paper presents a prototype called PADDIE-C19 (Physical Distancing Device with Edge Computing for COVID-19) to implement the physical distancing monitoring based on a low-cost edge computing device. The PADDIE-C19 provides real-time results and responses, as well as notifications and warnings to anyone who violates the 1-m physical distance rule. In addition, PADDIE-C19 includes temperature screening using an MLX90614 thermometer and ultrasonic sensors to restrict the number of people on specified premises. The Neural Network Processor (KPU) in Grove Artificial Intelligence Hardware Attached on Top (AI HAT), an edge computing unit, is used to accelerate the neural network model on person detection and achieve up to 18 frames per second (FPS). The results show that the accuracy of person detection with Grove AI HAT could achieve 74.65% and the average absolute error between measured and actual physical distance is 8.95 cm. Furthermore, the accuracy of the MLX90614 thermometer is guaranteed to have less than 0.5 °C value difference from the more common Fluke 59 thermometer. Experimental results also proved that when cloud computing is compared to edge computing, the Grove AI HAT achieves the average performance of 18 FPS for a person detector (kmodel) with an average 56 ms execution time in different networks, regardless of the network connection type or speed.

Download Full-text

ISEE: Industrial Internet of Things perception in solar cell detection based on edge computing

International Journal of Distributed Sensor Networks ◽

10.1177/15501477211050552 ◽

2021 ◽

Vol 17 (11) ◽

pp. 155014772110505

Author(s):

Meiya Dong ◽

Jumin Zhao ◽

Deng-ao Li ◽

Biaokai Zhu ◽

Sihai An ◽

...

Keyword(s):

Neural Network ◽

Power Generation ◽

Internet Of Things ◽

Real Time ◽

Edge Computing ◽

Theoretical Research ◽

Processing Unit ◽

Computing Unit ◽

Photovoltaic Power ◽

Industrial Internet

The photovoltaic industry is a strategic and sunrise industry with international competitive advantages. Driven by policy guidance and market demand, the new energy industry represented by the photovoltaic industry has been a significant emerging industry in developing the national economy and people’s livelihood. Stable photovoltaic power generation capacity supply is a critical issue in the photovoltaic industry. With the popularization of industrial Internet technology and Internet of things technology, more and more academic and industrial circles begin to introduce new technologies to provide the latest research results and solutions for the photovoltaic industry. Electroluminescence is a standard detection method for photovoltaic production in the application of solar energy production. This method uses human vision to detect whether the solar silicon unit is defective. In this article, due to the three core pain points in traditional electroluminescence detection: low efficiency of offline identification, low accuracy and accuracy of data detection, and no online diagnosis and prediction, we carry out ISEE research based on edge computing unit. ISEE uses the edge device to collect the real-time video image of the solar panel through the camera. Then it uses the powerful neural network processing unit module of the edge computing unit, combined with the convolutional neural network algorithm transplanted to the edge, to detect the defects of solar panels in real time. It completes the research on intelligent detection of photovoltaic power generation production defects based on the Internet of Things. After a large number of experimental design verification, ISEE effectively improves the automation degree and identification accuracy in the production and detection process of solar photovoltaic cells and reduces the cost of operation and maintenance. The accuracy rate reaches 93.75%, which has significant theoretical research significance and practical application value.

Download Full-text