Show simple item record

dc.contributor.authorPan, H.en_US
dc.contributor.authorBadawi, D.en_US
dc.contributor.authorZhang, X.en_US
dc.contributor.authorÇetin, Ahmet Enisen_US
dc.date.accessioned2020-02-04T08:19:17Zen_US
dc.date.available2020-02-04T08:19:17Zen_US
dc.date.issued2020en_US
dc.identifier.issn1863-1703
dc.identifier.urihttp://hdl.handle.net/11693/53035
dc.description.abstractIn this paper, we introduce a video-based wildfire detection scheme based on a computationally efficient additive deep neural network, which we call AddNet. This AddNet is based on a multiplication-free vector operator, which performs only addition and sign manipulation operations. In this regard, we construct a dot product-like operation from the mf-operator and use it to define dense and convolutional feed-forwarding passes in AddNet. We train AddNet on images taken from forestry surveillance cameras. Our experiments show that AddNet can achieve a time-saving by 12.4% when compared to an equivalent regular convolutional neural network (CNN). Furthermore, the smoke recognition performance of AddNet is as good as regular CNNs and substantially better than binary-weight neural networks.en_US
dc.language.isoEnglishen_US
dc.source.titleSignal, Image and Video Processingen_US
dc.relation.isversionofhttps://dx.doi.org/10.1007/s11760-019-01600-7en_US
dc.subjectComputationally efficienten_US
dc.subjectNeural networken_US
dc.subjectAdditive neural networken_US
dc.subjectReal-timeen_US
dc.subjectForest fire detectionen_US
dc.titleAdditive neural network for forest fire detectionen_US
dc.typeArticleen_US
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.citation.spage675en_US
dc.citation.epage682en_US
dc.citation.volumeNumber14en_US
dc.citation.issueNumber4en_US
dc.identifier.doi10.1007/s11760-019-01600-7en_US
dc.publisherSpringeren_US
dc.contributor.bilkentauthorÇetin, Ahmet Enisen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record