基于参数优化支持向量机的火焰图像识别

图像型火灾探测的主要问题是关于火焰和干扰物的识别。通过提取火灾图像特征,利用支持向量机来进行识别。为提高火灾准确预报率,用参数优化后的支持向量机来预报。提出一种混沌粒子群算法对支持向量机进行参数优化。实验表明,改进的粒子群算法比传统方法的火灾准确预报率有大幅提高,可以进一步降低火灾探测系统的误报。

Flame image recognition based on support vector machine parameter optimization

The main problem of image fire detection is the recognition for fire flame and interferences.Main characteristics of fire flame were obtained,and use support vector machine technology for image recognition.To improve the prediction accuracy rate of fire,the parameters optimized support vector machine to predict.This paper presents a chaotic particle swarm optimization algorithm for support vector machine parameter optimization.The experiment results show that the improved particle swarm optimization method in the prediction accuracy rate of fire has a substantial increase than that of the tradition,and can further reduce the false alarm of fire detection systems.