Utilize este identificador para referenciar este registo:
https://hdl.handle.net/10316/114621
Título: | Benchmarking Edge Computing Devices for Grape Bunches and Trunks Detection using Accelerated Object Detection Single Shot MultiBox Deep Learning Models | Autor: | Magalhães, Sandro Costa Santos, Filipe Neves Machado, Pedro Moreira, António Paulo Dias, Jorge |
Palavras-chave: | Embedded systems; Heterogeneous platforms; Object detection; SSD resNet; RetinaNet resNet | Data: | 21-Nov-2022 | Editora: | Elsevier | Projeto: | scholarship SFRH/BD/147117/2019 European Union’s Horizon 2020 Research and Innovation Program under Grant 101004085 |
Título da revista, periódico, livro ou evento: | Engineering Applications of Artificial Intelligence | Volume: | 117 | Resumo: | Purpose: Visual perception enables robots to perceive the environment. Visual data is processed using computer vision algorithms that are usually time-expensive and require powerful devices to process the visual data in real-time, which is unfeasible for open-field robots with limited energy. This work benchmarks the performance of different heterogeneous platforms for object detection in real-time. This research benchmarks three architectures: embedded GPU -- Graphical Processing Units (such as NVIDIA Jetson Nano 2 GB and 4 GB, and NVIDIA Jetson TX2), TPU -- Tensor Processing Unit (such as Coral Dev Board TPU), and DPU -- Deep Learning Processor Unit (such as in AMD-Xilinx ZCU104 Development Board, and AMD-Xilinx Kria KV260 Starter Kit). Method: The authors used the RetinaNet ResNet-50 fine-tuned using the natural VineSet dataset. After the trained model was converted and compiled for target-specific hardware formats to improve the execution efficiency. Conclusions and Results: The platforms were assessed in terms of performance of the evaluation metrics and efficiency (time of inference). Graphical Processing Units (GPUs) were the slowest devices, running at 3 FPS to 5 FPS, and Field Programmable Gate Arrays (FPGAs) were the fastest devices, running at 14 FPS to 25 FPS. The efficiency of the Tensor Processing Unit (TPU) is irrelevant and similar to NVIDIA Jetson TX2. TPU and GPU are the most power-efficient, consuming about 5W. The performance differences, in the evaluation metrics, across devices are irrelevant and have an F1 of about 70 % and mean Average Precision (mAP) of about 60 %. | URI: | https://hdl.handle.net/10316/114621 | ISSN: | 09521976 | DOI: | 10.1016/j.engappai.2022.105604 | Direitos: | openAccess |
Aparece nas coleções: | FCTUC Eng.Electrotécnica - Artigos em Revistas Internacionais |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
Benchmarking edge computing devices for grape bunches and trunks detection using accelerated object detection single shot multibox deep learning models.pdf | 4.47 MB | Adobe PDF | Ver/Abrir |
Visualizações de página
20
Visto em 15/mai/2024
Downloads
6
Visto em 15/mai/2024
Google ScholarTM
Verificar
Altmetric
Altmetric
Este registo está protegido por Licença Creative Commons