A New Wavelet Based SVM Classifier for Wild Fire Detection Using Decision Fusion Framework in Video

Abstract

There has been an increasing interest in the study of video based fire detection algorithms as video based surveillance systems become widely available for indoor and outdoor monitoring applications. Although many video based smoke-detection algorithms have been developed and applied in various experimental or real life applications, but the standard method for evaluating their quality has not yet been proposed. In this framework, it is assumed that the compound algorithm consists of several subalgorithms, each of which yields its own decision as a real number centered around zero, representing the confidence level of that particular subalgorithm. In this project, the wavelet support vector machine (WSVM)-based model is used for Wild fire detection (WFD). Decision values are linearly combined with weights that are updated online according to an active fusion method based on performing entropic projections onto convex sets describing subalgorithms. The new wavelet kernel is proposed to improve the generalization ability of the support vector machine (SVM). Moreover , the proposed model utilizes the principle of wavelet analysis to facilitate nonlinear characteristic extraction of the image data. To reduce misclassification due to fog, an efficient fog removal scheme using adaptive normalization method.