Fire Risk Assessment Using Neural Network and Logistic Regression
- PDF / 1,893,847 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 11 Downloads / 235 Views
RESEARCH ARTICLE
Fire Risk Assessment Using Neural Network and Logistic Regression Y. Jafari Goldarag 1 & Ali Mohammadzadeh 1
&
A. S. Ardakani 2
Received: 1 April 2015 / Accepted: 15 January 2016 # Indian Society of Remote Sensing 2016
Abstract Forest fire is one of the most important sources of land degradation that lead to deforestation and desertification processes. The presented work describes a methodology that employs logistic regression and artificial neural networks (ANN) to model forest fire risk and to recognize high potential area for fire occurrence. Different satellite and field data have been used in this work to model fire risk. These data include 12 static and dynamic parameters that are effective in fire occurrence and also 2001 to 2004 data was used to create model and data of year 2005 was used to evaluate created model. Two forest fire risk prediction models were created based on logistic regression and neural network in this research and both of them evaluated and compared. The result shows that neural network model is more accurate in fire point classification while logistic regression is sensitive to samples of fire points. To get high accuracy in logistic regression, it is necessary to be equilibrium the proportion of both fire and non-fire samples. Also different neural network structure was tested and the best architecture is a neural network with two hidden layer with 20, 28 neurons and logarithmic-sigmoid transfer function in both hidden layers. Accuracy of logistic regression and ANN in prediction of year 2005 fire was obtained 65.76 and 93.49, respectively. Keywords Neural network . Logistic regression . Fire . Forest . Remote sensing * Ali Mohammadzadeh [email protected]; [email protected]
1
Faculty of Geomatic Engineering, Remote Sensing Department, K.N. Toosi University of Technology, Vali_Asr St. Vanaq Sq., Mirdamad Cr., Tehran, Iran
2
GIS and Remote Sensing Department, Imam Hussein University, Tehran, Iran
Introduction A fire is a chemical reaction that its occurrence depends on heat, oxygen and fuel. These three factors form the Bfire fundamentals triangle^ (Pyne et al. 1996) and fire needs these to start and carry on. When an uncontrolled fire starts in natural vegetation, it converts to a forest fire (Bachmann and Allgower 2000). Generally, the beginning and expansion of a forest fire depends on the following factors: forest attribute, meteorological status, fire source, topographic conditions and so on. The fire danger level depends on the interaction between these factors (Huang et al. 2000). Forest fires are dangerous phenomenon and potentially a major threat in many regions of the world that would make a lot of losses in natural resources. Each year, fires affect almost 350 million hectares of land on the earth (Wright et al. 2002). Fires force changes to species composition and spatial pattern of vegetation cover and also can be a threat to human’s life and settlements. Fire would affect global changes and tropical ecosystems because of its connection to
Data Loading...
