DeepBrain: Experimental Evaluation of Cloud-Based Computation Offloading and Edge Computing in the Internet-of-Drones for Deep Learning Applications

Koubâa, Anis; Ammar, Adel; Alahdab, Mahmoud; Kanhouch, Anas; Azar, Ahmad Taher

doi:10.3390/s20185240

Cited by 38 publications

(13 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning [1] has emerged as a particular form of hierarchical learning that uses multi-layer neural networks to learn efficiently and progressively the useful data representations through the extraction of higher-level features from the raw input. It has demonstrated promising results and brought revolution in several applications related to the robotics [27] , object detection and recognition [28,29,30,31,32,33,34,35], image classification [36,37,38], character recognition [39,40], document classification [41], NLP [42] , speech recognition [39,41,43], human pose estimation [34], activity recognition [32,41], brain-computer interface [44], medical applications [45] and autonomous driving [46].…”

Section: A Edge Vs Cloud Computingmentioning

confidence: 99%

Edge Deployment Framework of GuardBot for Optimized Face Mask Recognition With Real-Time Inference Using Deep Learning

Manzoor¹,

Kim

Joo

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Deep learning based models on the edge devices have received considerable attention as a promising means to handle a variety of AI applications. However, deploying the deep learning models in the production environment with efficient inference on the edge devices is still a challenging task due to computation and memory constraints. This paper proposes a framework for the service robot named GuardBot powered by Jetson Xavier NX and presents a real-world case study of deploying the optimized face mask recognition application with real-time inference on the edge device. It assists the robot to detect whether people are wearing a mask to guard against COVID-19 and gives a polite voice reminder to wear the mask. Our framework contains dual-stage architecture based on convolutional neural networks with three main modules that employ (1) MTCNN for face detection, (2) our proposed CNN model and seven transfer learning based custom models which are Inception-v3, VGG16, denseNet121, resNet50, NASNetMobile, XceptionNet, MobileNet-v2 for face mask classification, (3) TensorRT for optimization of all the models to speedup inference on the Jetson Xavier NX. Our study carries out several analysis based on the models' performance in terms of their frames per second, execution time and images per second. It also evaluates the accuracy, precision, recall & F1-score and makes the comparison of all models before and after optimization with a main focus on high throughput and low latency. Finally, the framework is deployed on a mobile robot to perform experiments in both outdoor and multi-floor indoor environments with patrolling and non-patrolling modes. Compared to other state-of-the-art models, our proposed CNN model for face mask recognition based on the classification obtains 94.5%, 95.9% and 94.28% accuracy on training, validation and testing datasets respectively which is better than MobileNet-v2, Xception and InceptionNet-v3 while it achieves highest throughput and lowest latency than all other models after optimization at different precision levels.

show abstract

Section: A Edge Vs Cloud Computingmentioning

confidence: 99%

Edge Deployment Framework of GuardBot for Optimized Face Mask Recognition With Real-Time Inference Using Deep Learning

Manzoor¹,

Kim

Joo

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Data submission to the cloud or the use of a bigger drone can reduce energy consumption for data processing, but the computation offloading, and wireless data transmission will both be sacrificed. Energy management strategies for mobile devices from hardware to software aspects, especially on drones are the focus of many studies, such as [137], which proposes a system architecture for inter-connected drones to allow for the usage of deep learning on edge and on the cloud. The authors in [138,139] propose other architectures based on blockchain technology to demonstrate the prolonged drone operating time, but conclude that efficiency and reduced consumption are still not sufficient for rescue and search operations.…”

Section: Power Consumption and Flight Timementioning

confidence: 99%

Ionizing Radiation Monitoring Technology at the Verge of Internet of Things

Ahmad

Rahim²,

Nordin

et al. 2021

Sensors

View full text Add to dashboard Cite

As nuclear technology evolves, and continues to be used in various fields since its discovery less than a century ago, radiation safety has become a major concern to humans and the environment. Radiation monitoring plays a significant role in preventive radiological nuclear detection in nuclear facilities, hospitals, or in any activities associated with radioactive materials by acting as a tool to measure the risk of being exposed to radiation while reaping its benefit. Apart from in occupational settings, radiation monitoring is required in emergency responses to radiation incidents as well as outdoor radiation zones. Several radiation sensors have been developed, ranging from as simple as a Geiger-Muller counter to bulkier radiation systems such as the High Purity Germanium detector, with different functionality for use in different settings, but the inability to provide real-time data makes radiation monitoring activities less effective. The deployment of manned vehicles equipped with these radiation sensors reduces the scope of radiation monitoring operations significantly, but the safety of radiation monitoring operators is still compromised. Recently, the Internet of Things (IoT) technology has been introduced to the world and offered solutions to these limitations. This review elucidates a systematic understanding of the fundamental usage of the Internet of Drones for radiation monitoring purposes. The extension of essential functional blocks in IoT can be expanded across radiation monitoring industries, presenting several emerging research opportunities and challenges. This article offers a comprehensive review of the evolutionary application of IoT technology in nuclear and radiation monitoring. Finally, the security of the nuclear industry is discussed.

show abstract

“…Koubaa et al [17] introduced an IoD architecture for computation offloading. The architecture with drones, edge, and cloud allows for offloading tasks from drones over the Internet to cloud servers powerful in computation and memory storage.…”

Section: Related Workmentioning

confidence: 99%

A Novel Internet-of-Drones and Blockchain-based System Architecture for Search and Rescue

Nguyen¹,

Katila

Nguyen

2021

2021 IEEE 18th International Conference on Mobile Ad Hoc and Smart Systems (MASS)

View full text Add to dashboard Cite

With the development in information and communications technology (ICT) and drones such as Internet-of-Things (IoT), edge computing, image processing, and autonomous drones, solutions supporting search and rescue (SAR) missions can be developed with more intelligent capabilities. In most of the drone and unmanned aerial vehicle (UAV) based systems supporting SAR missions, several drones deployed in different areas acquire images and videos that are sent to a ground control station (GCS) for processing and detecting a missing person. Although this offers many advantages, such as easy management and deployment, the approach still has many limitations. For example, when a connection between a drone and a GCS has some problems, the quality of service cannot be maintained. Many drone and UAV-based systems do not support flexibility, transparency, security, and traceability. In this paper, we propose a novel Internet-of-Drones (IoD) architecture using blockchain technology. We implement the proposed architecture with different drones, edge servers, and a Hyperledger blockchain network. The proof-of-concept design demonstrates that the proposed architecture can offer high-level services such as prolonging the operating time of a drone, improving the capability of detecting humans accurately, and a high level of transparency, traceability, and security.

show abstract

DeepBrain: Experimental Evaluation of Cloud-Based Computation Offloading and Edge Computing in the Internet-of-Drones for Deep Learning Applications

Cited by 38 publications

References 40 publications

Edge Deployment Framework of GuardBot for Optimized Face Mask Recognition With Real-Time Inference Using Deep Learning

Edge Deployment Framework of GuardBot for Optimized Face Mask Recognition With Real-Time Inference Using Deep Learning

Ionizing Radiation Monitoring Technology at the Verge of Internet of Things

A Novel Internet-of-Drones and Blockchain-based System Architecture for Search and Rescue

Contact Info

Product

Resources

About