作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (19): 166-167,176. doi: 10.3969/j.issn.1000-3428.2011.19.054

• 人工智能及识别技术 • 上一篇    下一篇

基于特征融合的图像型火灾探测方法

王媛彬,马宪民   

  1. (西安科技大学电气与控制工程学院,西安 710054)
  • 收稿日期:2011-04-12 出版日期:2011-10-05 发布日期:2011-10-05
  • 作者简介:王媛彬(1977-),女,讲师、硕士,主研方向:图像处理,模式识别;马宪民,教授、博士
  • 基金资助:
    国家自然科学基金资助项目(50977077);国家科技部专 项基金资助项目(2009GJG00020);陕西省教育厅科研计划基金资助项目(11JK0908)

Image Fire Detection Method Based on Feature Fusion

WANG Yuan-bin, MA Xian-min   

  1. (School of Electrical and Control Engineering, Xi’an University of Science and Technology, Xi’an 710054, China)
  • Received:2011-04-12 Online:2011-10-05 Published:2011-10-05

摘要: 针对传统火灾探测中灵敏度不高、响应慢的问题,提出一种基于特征融合的图像型火灾探测方法。结合火焰的颜色、运动以及闪烁特征,检测出疑似火灾区域中的火焰像素,排除非火焰像素,并用支持向量机对疑似火焰像素进行验证,采用形态学方法和区域融合判断出火灾区域。实验结果表明,该方法对多种火灾和非火灾场景具有较好的适应性、较强的抗干扰能力以及较高的探测率。

关键词: 火灾探测, 动态特征, 静态特征, 支持向量机, 特征融合, 图像处理

Abstract: In order to overcome the disadvantages of traditional fire detection, such as low sensitivity and speed, an image fire detection method based on feature fusion is proposed. According to the color, motion and flame flicker features, fire pixel in candidate fire region is detected and non-fire pixels are removed. The candidate flame pixels are verified by support vector machines with a radial basis function kernel. Fire pixels which are verified are merged into ?re regions by using morphological closing and region. Experimental results show that the proposed method can recognize fire effectively and has higher accuracy. It is more robust to noise, strong anti-jamming and high detection rate.

Key words: fire detection, dynamic feature, static feature, Support Vector Machine(SVM), feature fusion, image processing

中图分类号: