Document


Title

Additive neural network for forest fire detection
Document Type: Journal Article
Author(s): Hongyi Pan; Diaa Badawi; Xi Zhang; Ahmet Enis Cetin
Publication Year: 2020

Cataloging Information

Keyword(s):
  • additive neural network
  • computationally efficient
  • fire detection
  • neural network
  • real time
Record Maintained By:
Record Last Modified: May 14, 2020
FRAMES Record Number: 60517

Description

In 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.

Online Link(s):
Citation:
Pan, Hongyi; Badawi, Diaa; Zhang, Xi; Cetin, Ahmet Enis. 2020. Additive neural network for forest fire detection. Signal, Image and Video Processing 14(4):675-682.