Forest fire image recognition based on convolutional neural network

In order to detect fire automatically, a forest fire image recognition method based on convolutional neural networks is proposed in this paper. There are two main types of fire recognition algorithms. One is based on traditional image processing technology and the other is based on convolutional neu...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of algorithms & computational technology Ročník 13
Hlavní autoři: Wang, Yuanbin, Dang, Langfei, Ren, Jieying
Médium: Journal Article
Jazyk:angličtina
Vydáno: London, England SAGE Publications 01.11.2019
Sage Publications Ltd
SAGE Publishing
Témata:
Algorithms
Artificial neural networks
Blindness
Feature extraction
Forest fires
Image detection
Image processing
Neural networks
Object recognition
deep learning
Forest fire recognition
pooling model
convolutional neural network
ISSN:1748-3018, 1748-3026
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In order to detect fire automatically, a forest fire image recognition method based on convolutional neural networks is proposed in this paper. There are two main types of fire recognition algorithms. One is based on traditional image processing technology and the other is based on convolutional neural network technology. The former is easy to lead in false detection because of blindness and randomness in the stage of feature selection, while for the latter the unprocessed convolutional neural network is applied directly, so that the characteristics learned by the network are not accurate enough, and recognition rate may be affected. In view of these problems, conventional image processing techniques and convolutional neural networks are combined, and an adaptive pooling approach is introduced. The fire flame area can be segmented and the characteristics can be learned by this algorithm ahead. At the same time, the blindness in the traditional feature extraction process is avoided, and the learning of invalid features in the convolutional neural network is also avoided. Experiments show that the convolutional neural network method based on adaptive pooling method has better performance and has higher recognition rate.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1748-3018
1748-3026
DOI:10.1177/1748302619887689