A Real-time Fire Segmentation Method Based on A Deep Learning Approach. Issue 6 (2022)

• Record Type: Journal Article
• Authors: Li, Mengna; Zhang, Youmin; Mu, Lingxia; Xin, Jing; Yu, Ziquan; Jiao, Shangbin; Liu, Han; Xie, Guo; Yingmin, Yi
• Is Part Of: IFAC-PapersOnLine. Volume 55:Issue 6(2022)
• Page Start: 145
• Page End: 150
• Publication Date: 2022
• Subjects: Forest fire -- Fire segmentation -- Atrous spatial pyramid pooling -- Mobilenetv3 -- Deeplabv3+; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2022.07.120
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: As a kind of the forest "fault", fire is highly destructive and difficult to rescue. Fire segmentation is helpful for firefighters to understand the fire scale and formulate a reasonable fire-fighting plan. Therefore, this paper proposes a real-time fire segmentation method based on deep learning. This method is an improved version of deeplbav3+, which is an encoder-decoder structure network. Encoder network is composed of deep convolutional neural network and atrous spatial pyramid pooling. Different from deeplabv3+, in order to improve the segmentation speed, this paper uses the lightweight network mobilenetv3 to build a new deep convolutional neural network and does not use atrous convolution, but it will affect the segmentation accuracy. Therefore, in order to compensate for the loss of segmentation accuracy, on the basis of the original decoder network, this paper adds two different shallow features to make the network contain rich fire feature information. Experimental results show that the comprehensive performance of this method is better than the original deeplabv3+, especially the segmentation speed of the network is greatly improved, which is about 59 FPS.