【摘要】 根据火焰的颜色特征提出一种基于OHTA颜色空间并利用Otsu阈值选取来分割火焰区域的方法。该方法能够在室外不同环境下准确地分割出含有火焰的区域。结合火焰的动态特征,利用三帧差分法进一步去掉具有火焰颜色的静止物体。该方法时间复杂度低,能够满足视频检测的实时性要求。更多还原

【Abstract】 According to the characteristics of flame colour,we propose a new method,it is based on OHTA colour space and uses Ostu threshold to segment flame area. The method can accurately segment the area containing the flame in different environment outdoors. Combining with dynamic characteristics of the flame,we use three-frame difference method to further remove the stationary object with flame colour. The method has low time complexity,can meet the real-time requirements of video detection.更多还原