Memory-Based Pruning of Deep Neural Networks for IoT Devices Applied to Flood Detection

Autor(es)

Fernandes Junior, Francisco Erivaldo ^[1] ; Nonato, Luis Gustavo ^[2] ; Ranieri, Caetano Mazzoni ^[2] ; Ueyama, Jo ^[2]

Afiliação do(s) autor(es)

^[1] SIDIA R&D Inst, BR-69055035 Manaus, Amazonas - Brazil

^[2] Univ Sao Paulo, Inst Math & Comp Sci, BR-13566590 Sao Carlos - Brazil

• Tipo de documento: Artigo Científico
• Fonte: SENSORS; v. 21, n. 22 NOV 2021.
• Citações Web of Science: 0

Resumo

Automatic flood detection may be an important component for triggering damage control systems and minimizing the risk of social or economic impacts caused by flooding. Riverside images from regular cameras are a widely available resource that can be used for tackling this problem. Nevertheless, state-of-the-art neural networks, the most suitable approach for this type of computer vision task, are usually resource-consuming, which poses a challenge for deploying these models within low-capability Internet of Things (IoT) devices with unstable internet connections. In this work, we propose a deep neural network (DNN) architecture pruning algorithm capable of finding a pruned version of a given DNN within a user-specified memory footprint. Our results demonstrate that our proposed algorithm can find a pruned DNN model with the specified memory footprint with little to no degradation of its segmentation performance. Finally, we show that our algorithm can be used in a memory-constraint wireless sensor network (WSN) employed to detect flooding events of urban rivers, and the resulting pruned models have competitive results compared with the original models. (AU)

• Processo FAPESP: 13/07375-0 - CeMEAI - Centro de Ciências Matemáticas Aplicadas à Indústria
• Beneficiário: Francisco Louzada Neto
• Modalidade de apoio: Auxílio à Pesquisa - Centros de Pesquisa, Inovação e Difusão - CEPIDs
• Processo FAPESP: 20/05426-0 - Usando processamento de imagens para detectar e prever enchentes
• Beneficiário: Francisco Erivaldo Fernandes Junior
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado