Moving target detection of elevating fire truck based on D-vibe algorithm in vehicular environment
Zhen Yuan, Degang Xu · 2022
At present, fire trucks are widely used in various fire fighting and emergency tasks. In order to deal with different situation, fire trucks are equipped with several different rescue tools. It is very important for quickly replace tools of fire trucks for different rescue purpose. In this paper machine vision is utilized for Intelligent recognition and positioning the tools on the long mechanical arm in the process of replacing the tools when the mechanical arm moving in vehicle environment. We acquired the RGB-D image including the pixel mapping relationship between the color image and the depth image. Then filtering algorithms are used to process the depth image. A D-vibe algorithm based on depth image is proposed to get the characteristics of depth image. The spatiotemporal depth information is introduced to construct the depth model for Intelligent recognition and positioning. Experimental results show that our method can effectively eliminate the "Edge Ghost" caused by shaking video and improve the target detection effect.