基于多特征融合的早期野火烟雾检测

An Early Wildfire Smoke Detection Method based on Multi-features Fusion

随着计算机技术的不断发展,基于计算机视觉和模式识别的森林火灾烟雾检测算法具有很大的应用前景。针对目前检测方法适应性不强、在复杂环境下检测识别率不高的问题,提出一种通过融合烟雾多个特征的方法来检测识别早期林火烟雾。算法首先通过一种结合改进的四帧差分法和高斯混合背景建模的算法提取运动前景;然后利用烟雾颜色特征、小波变换分析和LBP纹理特征,利用多特征线性融合并通过K最近邻(KNN)分类算法进行识别。通过在不同视频场景中的实验,证明了该方法在烟雾检测能力上的有效性。

With the development of computer technology,video-based forest fire smoke detection algorithms based on computer vision and pattern recognition have great application prospects. Because the current detection methods are not flexible and recognition rate is not high,this paper proposed a novel wildfire smoke detection algorithm based on multifeatures fusion. Firstly,the algorithm extracts the motion foreground by an improved four-frame difference method and Gaussian Mixture Model. Then,the linear combination of smoke color features,wavelet transform analysis and LBP texture features are used to identify the video by multi-feature linear fusion and KNN classifier. Experiments in different video scenes verify the effectiveness of the proposed method in smoke detection.