Smart Surveillance as an Edge Network Service: From Harr-Cascade, SVM to a Lightweight CNN

Nikouei, Seyed Yahya; Song, Sejun; Xu, Ronghua; Choi, Baek‐Young; Faughnan, Timothy R.

doi:10.1109/cic.2018.00042

Cited by 73 publications

(43 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results showed that the lower number of CNN gave a lower computational overhead, then decreased the computation time of the image processing. Nikouei et al [25,26] proposed a lightweight convolutional neural network (L-CNN) by using the depth wise separable convolution. The aim of the L-CNN was to reduce the number of parameters in the network without affecting the quality of the output so that it could fit with edge devices such as Raspberry Pi.…”

Section: Related Workmentioning

confidence: 99%

Low-Cost Image Search System on Off-Line Situation

et al. 2020

View full text Add to dashboard Cite

Implementation of deep learning in low-cost hardware, such as an edge device, is challenging. Reducing the complexity of the network is one of the solutions to reduce resource usage in the system, which is needed by low-cost system implementation. In this study, we use the general average pooling layer to replace the fully connected layers on the convolutional neural network (CNN) model, used in the previous study, to reduce the number of network properties without decreasing the model performance in developing image classification for image search tasks. We apply the cosine similarity to measure the characteristic similarity between the feature vector of image input and extracting feature vectors from testing images in the database. The result of the cosine similarity calculation will show the image as the result of the searching image task. In the implementation, we use Raspberry Pi 3 as a low-cost hardware and CIFAR-10 dataset for training and testing images. Base on the development and implementation, the accuracy of the model is 68%, and the system generates the result of the image search base on the characteristic similarity of the images.

show abstract

Section: Related Workmentioning

confidence: 99%

Low-Cost Image Search System on Off-Line Situation

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Implemented based on the edge-fog-cloud hierarchy architecture. the input frame that is streamed out of the surveillance camera is given to an edge unit where low-level processing is performed [21], [22]. The intermediate-level is fog nodes, where multiple tasks are performed based on the processing power and resources available.…”

Section: A Smart Public Safetymentioning

confidence: 99%

BlendMAS: A Blockchain-Enabled Decentralized Microservices Architecture for Smart Public Safety

Nikouei

Chen

et al. 2019

2019 IEEE International Conference on Blockchain (Blockchain)

Self Cite

View full text Add to dashboard Cite

Thanks to rapid technological advances in the Internet of Things (IoT), a smart public safety (SPS) system has become feasible by integrating heterogeneous computing devices to collaboratively provide public protection services. While a service oriented architecture (SOA) has been adopted by IoT and cyberphysical systems (CPS), it is difficult for a monolithic architecture to provide scalable and extensible services for a distributed IoT based SPS system. Furthermore, traditional security solutions rely on a centralized authority, which can be a performance bottleneck or single point failure. Inspired by microservices architecture and blockchain technology, this paper proposes a BLockchain-ENabled Decentralized Microservices Architecture for Smart public safety (BlendMAS). Within a permissioned blockchain network, a microservices based security mechanism is introduced to secure data access control in a SPS system. The functionality of security services are decoupled into separate containerized microservices that are built using a smart contract, and deployed on edge and fog computing nodes. An extensive experimental study verified that the proposed BlendMAS is able to offer a decentralized, scalable and secured data sharing and access control to distributed IoT based SPS system.

show abstract

“…A. PReS System Level View Figure 3 shows the proposed PReS system following a hierarchical architecture [2], [9]. At the sensor level, multiple data are collected by the sensors to obtain the real-time status of the train, including the GPS position, the line section, speed, power system status, etc.…”

Section: Real-time Scheduling For Power Peak Reductionmentioning

confidence: 99%

PReS: Power Peak Reduction by Real-time Scheduling for Urban Railway Transit

Yang

Chen

Zhou

et al. 2019

2019 IEEE Sustainability Through ICT Summit (StICT)

Self Cite

View full text Add to dashboard Cite

Railway transportation is one of the most popular options for Urban Massive Transportation Systems (UMTS) because of many attractive features. A robust electric power supply is essential to enable normal operation. However, the power peaks appearing at the start time of the vehicles put heavy pressure on the power grid. Reduction of the power peak is a key issue in improving urban railway transit's power efficiency. Researchers have tried to address this problem by making a delicate timetable, but this method often failed to serve the purpose because of the punctuality problem. In this work, taking advantage of real-time estimation of the single train's power consumption, an online Power peak Reduction by real-time Scheduling (PReS) solution for trains' departure is proposed. Particularly, a Binary Integer Programming (BIP) model is introduced that is able to avoid power consumption peak caused by multiple trains departure simultaneously. The simulation result verified that the proposed real-time scheduling approach can effectively reduce the occurrences of power peak without bringing in additional train travel delay.

show abstract

Smart Surveillance as an Edge Network Service: From Harr-Cascade, SVM to a Lightweight CNN

Cited by 73 publications

References 29 publications

Low-Cost Image Search System on Off-Line Situation

Low-Cost Image Search System on Off-Line Situation

BlendMAS: A Blockchain-Enabled Decentralized Microservices Architecture for Smart Public Safety

PReS: Power Peak Reduction by Real-time Scheduling for Urban Railway Transit

Contact Info

Product

Resources

About