Distributed Deep Learning in the Cloud and Energy-efficient Real-time Image Processing at the Edge for Fish Segmentation in Underwater Videos Segmentation in Underwater Videos

Jahanbakht, Mohammad, Xiang, Wei, Waltham, Nathan J., and Rahimiazghadi, Mostafa (2022) Distributed Deep Learning in the Cloud and Energy-efficient Real-time Image Processing at the Edge for Fish Segmentation in Underwater Videos Segmentation in Underwater Videos. IEEE Access, 10. pp. 117796-117807.

Preview

PDF (Accepted Publisher Version) - Accepted Version
Available under License Creative Commons Attribution.
Download (1MB) | Preview

DOI: 10.1109/ACCESS.2022.3202975

View at Publisher Website: https://doi.org/10.1109/ACCESS.2022.3202...

639

Abstract

Using big marine data to train deep learning models is not efficient, or sometimes even possible, on local computers. In this paper, we show how distributed learning in the cloud can help more efficiently process big data and train more accurate deep learning models. In addition, marine big data is usually communicated over wired networks, which if possible to deploy in the first place, are costly to maintain. Therefore, wireless communications dominantly conducted by acoustic waves in underwater sensor networks, may be considered. However, wireless communication is not feasible for big marine data due to the narrow frequency bandwidth of acoustic waves and the ambient noise. To address this problem, we propose an optimized deep learning design for low-energy and real-time image processing at the underwater edge. This leads to trading the need to transmit the large image data, for transmitting only the low-volume results that can be sent over wireless sensor networks. To demonstrate the benefits of our approaches in a real-world application, we perform fish segmentation in underwater videos and draw comparisons against conventional techniques. We show that, when underwater captured images are processed at the collection edge, 4 times speedup can be achieved compared to using a landside server. Furthermore, we demonstrate that deploying a compressed DNN at the edge can save 60% of power compared to a full DNN model. These results promise improved applications of affordable deep learning in underwater exploration, monitoring, navigation, tracking, disaster prevention, and scientific data collection projects.


Item ID:	75959
Item Type:	Article (Research - C1)
ISSN:	2169-3536
Keywords:	Big marine data, Deep learning, Edge computing, Fish image segmentation, Internet of underwater things, Real-time video processing, U-Net convolutional neural networks.
Related URLs:	https://researchonline.jcu.edu.au/76919/
Copyright Information:	This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
Date Deposited:	12 Sep 2022 22:47
FoR Codes:	41 ENVIRONMENTAL SCIENCES > 4104 Environmental management > 410401 Conservation and biodiversity @ 30% 41 ENVIRONMENTAL SCIENCES > 4104 Environmental management > 410404 Environmental management @ 30% 31 BIOLOGICAL SCIENCES > 3103 Ecology > 310305 Marine and estuarine ecology (incl. marine ichthyology) @ 40%
SEO Codes:	18 ENVIRONMENTAL MANAGEMENT > 1802 Coastal and estuarine systems and management > 180203 Coastal or estuarine biodiversity @ 50% 18 ENVIRONMENTAL MANAGEMENT > 1805 Marine systems and management > 180507 Rehabilitation or conservation of marine environments @ 50%
Downloads:	Total: 639 Last 12 Months: 13
	More Statistics

Actions (Repository Staff Only)

Item Control Page