摘要
针对煤矿井下复杂环境中传统火灾探测的不足,提出了基于图像处理技术的SVM分类器模型进行火灾探测。首先利用颜色空间进行火焰图像的分割,进而提取出尖角数、圆形度、质心位移等特征,最后通过改进人工蜂群算法优化参数后的支持向量机融合特征量进行分类,以取得最优的探测效果。
Prediction SVM classifier model for fire in coal mine based on image processing technology was proposed according to the impacts of environmental factors to the traditional fire detection in coal mine. Firstly, using image segmentation based on the flame color space, and then extract the number of sharp corners, roundness, displacement and other features, optimization parameters by artificial bee colony after SVM classification integration of these features amount to achieve optimal classification.
出处
《煤炭技术》
CAS
北大核心
2016年第4期187-189,共3页
Coal Technology
