A 3d feature is extracted from the accumulation and main motionorientation, and a Bayesian classifier is used for smoke detection.
根据累积量和主运动方向提取出3维特征矢量,采用贝叶斯分类器进行烟雾的检测。
2
In order to accelerate detection speed, video images are divided into blocks. Each block motionorientation is estimated by block matching methods.
为了加快检测速度,将视频图像分割成大小相等的块,并估计每个块的运动方向。
3
And a time sequence of motionorientation for each block is generated over a sliding time window. Then accumulation and main motionorientation are computed according to the sequence.