.png){kind=icon}
.png){kind=icon}
Return
EdgeEye: A Long-Range Energy-Efficient Vision Node For Long-Term Edge Computing
.png){kind=icon}
PRE
AI
DOI：10.1109/igsc48788.2019.8957170.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recent Internet of Things (IoT) devices start to integrate data-intensive sensors such as cameras or microphones. To reduce the amount of data that needs to be transmitted to the cloud, an IoT device needs to process the data on-board, which is commonly referred to as edge computing. This paper presents a stand-alone, edge computing device called EdgeEye, capable of data-centric processing in the mW range. It employs a GAP8 processor which belongs to the PULP processor family to perform machine learning inference on images that were acquired with an ultra-low-power (ULP) camera. The results of the inference are transmitted using long-range communication (LoRa and LoRaWAN). EdgeEye is evaluated in the application scenario of people counting, using a newly proposed convolutional neural network that runs in the 8-cores of GAP8. The system performs the people recognition task (image acquisition, image processing, and LoRa transmission) consuming only 17.5 mJ. The estimated system battery life is more than 400 days when performing a new prediction once per minute on a 2000 mAh battery (3.7 V).
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
I
IF:
0
Papers:
8
Citations:
0