基于视频数据的电梯门防夹检测算法研究

Research on Elevator Door Anti-pinch Detection Algorithm Based on Video Data

针对电梯门现有防夹保护装置的缺陷,提出了一种基于视频数据的防夹检测算法,用于在轿厢门区内检测物体;首先,采用改进的双边滤波方法滤除视频帧和背景图像中的噪声;然后,差分视频帧与背景图像,并采用极大熵粒子群算法来二值化差分图像;接着,将二值图像进行形态学滤波以检测出轿厢门区范围内的物体;最后,利用Surendra算法更新背景图像;实验结果表明,提出算法可以在轿厢门区范围内实时且准确地检测出物体,为电梯门防夹功能的改进提供了可靠的保障。

For the defects of the existing anti-pinch protection device,an anti-pinch detection algorithm is proposed to detect objects in the elevator door area.First,an improved bilateral filtering method is used to remove the noise of the video frame and background image.Then,a difference image of the video frame and the background image is calculated,and the maximum entropy particle swarm optimization algorithm is employed to obtain binary image of the difference image.Next,objects in the door area are detected from the binary image by morphology processing.Final,the Surendra algorithm is adopted to update the background image.Experimental results show that the proposed algorithm can detect objects accurately in the elevator door area.It provides a reliable guarantee for the improved anti-pinch function of elevator door.