计算机应用 ›› 2013, Vol. 33 ›› Issue (03): 704-707.DOI: 10.3724/SP.J.1087.2013.00704

• 多媒体处理技术 • 上一篇    下一篇

基于粗糙集的火灾图像特征选择与识别

胡燕1,2*,王慧琴1,2,秦薇薇2,邹婷2,梁俊山2   

  1. 1.西安建筑科技大学 管理学院, 西安 710055;
    2.西安建筑科技大学 信息与控制工程学院, 西安 710055
  • 收稿日期:2012-09-14 修回日期:2012-11-13 出版日期:2013-03-01 发布日期:2013-03-01
  • 通讯作者: 胡燕
  • 作者简介:胡燕(1981-),女,河南杞县人,工程师,博士研究生,主要研究方向:信息安全、数字图像处理; 王慧琴(1970-),女,山西长治人,教授,博士生导师,博士,主要研究方向:数字图像处理、计算机与通信网络安全、智能信息处理; 秦薇薇(1987-),女,河北邯郸人,硕士研究生,主要研究方向:数字图像处理; 邹婷(1985-),女,陕西西安人,硕士研究生,主要研究方向:数字图像处理; 梁俊山(1981-),男,河北邯郸人,硕士研究生,主要研究方向:数字图像处理。
  • 基金资助:

    陕西省教育厅产业化项目(2011JG12); 榆林市科技计划项目; 西安建筑科技大学青年科技基金资助项目(QN1125)。

Fire image features selection and recognition based on rough set

HU Yan1,2*, WANG Huiqin1,2, QIN Weiwei2, ZOU Ting2, LIANG Junshan2   

  1. 1. School of Management, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China;
    2. School of Information and Control Engineering, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China
  • Received:2012-09-14 Revised:2012-11-13 Online:2013-03-01 Published:2013-03-01

摘要: 针对图像型火灾探测方法检测准确度和实时性间的矛盾,提出了基于粗糙集的火灾图像特征选择和识别算法。首先通过对火焰图像特征的深入研究发现,在燃烧能量的驱动下火焰的上边缘极不规则,出现明显的震动现象,而下边缘却恰恰相反; 基于此特点,可利用上下边缘抖动投影个数比作为火焰区别于边缘形状较规则的干扰。然后,选择火焰的6个显著特征构造训练样本,在火灾分类能力不受影响的前提下,使用实验所得的特征量归类表对训练样本进行属性约简,并将约简后的信息系统属性训练支持向量机模型,实现火灾探测。最后与传统支持向量机火灾探测算法做了比较。实验结果表明:将粗糙集作为支持向量机分类器的前置系统,把粗糙集理论的属性约简引入到支持向量机中,可以大大消除样本集冗余属性,降低了火灾图像特征空间的维数,减少了分类器训练和检测数据,在保证识别精度的同时,提高了算法的速度和泛化能力。

关键词: 火灾图像特征, 粗糙集, 属性约简, 支持向量机, 火灾识别

Abstract: Concerning the contradiction of accuracy and real-time in image fire detection, a fire image features selection and recognition algorithm based on rough set was proposed. Firstly, through in-depth study on the flame image features, the top edge of flame driven by the combustion energy is very irregular, and obvious vibration phenomenon occurres. But the lower edge is the opposite. Based on this feature, the upper and lower edges of the jitter projection ratio can be used as a flame from the edge shape regular interference. Then, the six striking flame features were chosen in order to create training samples. When fire classification ability was not affected, the feature classification table gained by experiment was used to reduce attributes of the training samples. And the reduced information systems attributes were applied to train a support vector machine model, and the fire detection was realized. Finally, this fire detection algorithm was compared to the traditional Support Vector Machine (SVM) fire detection algorithm. The results show that the presented algorithm reduces redundant attributes, eliminates the dimension of fire image features space, and decreases the data of training and testing in classifier in case rough set as a SVM classifier prefix system. While ensuring recognition accuracy, the algorithm improves fire detection speed.

Key words: fire image feature, Rough Set (RS), Attribute Reduction (AR), Support Vector Machine (SVM), fire recognition

中图分类号: