.png){kind=icon}
.png){kind=icon}
Return
Wildfire detection using modified particle swarm optimization algorithm
.png){kind=icon}
PRE
AI
DOI：10.47974/jios-2303.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Wildfires claim many lives yearly and cause huge environmental and economic damage. Some of the worst wildfire disasters can be found in world history. Reducing CO2 emissions is critical to prevent irreversible change in the world climate in the era of global warming. Wildfire is the primary factor that leads to high CO2 emissions, promoting global warming. Early detection and prevention of wildfires are essential tasks of computer vision. An automated wildfire classification system can help prevent wildfires at an early stage. Many methods are used to identify wildfires and automate categorizing wildfires based on images. This paper offers a new path to wildfire classification based on a wildfire image dataset. This paper introduces a modified particle swarm optimization algorithm for classifying forest fire images using the image dimensions and characteristics that identify forest fires. The proposed approach uses a dataset to identify forest fires. Whether it is classified as a fire is based on the images of fire and smoke in the air. The proposed approach used an image dataset to validate and test the algorithm and outperform it in terms of precision, recall, F-score, and accuracy.
Keywords:
Particle swarm optimization
Engineering optimization
Nature inspired algorithm
Natural disaster
Journal
J
IF:
0.7
Papers:
128
Citations:
0