Name
Abstract | ||
|---|---|---|
A fire detection framework based on image processing is presented in this paper. The proposed framework incorporates Scale-Invariant Feature Transform (SIFT) features and applies it in a novel way for use in fire detection by taking advantage of SIFT's ability to learn and adapt itself with various datasets. The framework was connected to a number of clusters and classifiers and was trained and tested with several fire and non fire image datasets. The performance of two classifiers in terms of the accuracy and sensitivity was examined and a comparison between the proposed framework and an existing image processing fire detection method has been presented. The experimental results, using the Support Vector Machine (SVM) classification, show that the proposed framework using SIFT features performs well and can achieve an accuracy of 94.7%. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/ICSPCS.2018.8631711 | ICSPCS |
Keywords | Field | DocType |
Feature extraction,Forestry,Clustering algorithms,Support vector machines,Visualization,Image color analysis,Histograms | Scale-invariant feature transform,Histogram,Pattern recognition,Computer science,Visualization,Support vector machine,Image processing,Feature extraction,Artificial intelligence,Cluster analysis,Fire detection | Conference |
ISBN | Citations | PageRank |
978-1-5386-5602-0 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Nargess Ghassempour | 1 | 0 | 0.34 |
| Ju Jia Zou | 2 | 198 | 20.00 |
| Yaping He | 3 | 0 | 0.34 |