Fast and intelligent measurement of concrete aggregate volume based on monocular vision mapping

In order to prevent the abnormal appearance of gravel aggregate material level in the mixing plant, improve the safety of the concrete mixing plant system as well as the efficient and high-quality production of the concrete mixing plant, this paper proposes a monocular vision-based fast and intelligent measurement method for the volume of concrete aggregate. This method combines peak and valley positioning information derived from target detection model to find the height of aggregate level and diameter of the surface circle formed by the valley. The relationship between this localization data and the height, angle and volume data are then analyzed separately, and finally, the volume of the aggregate is calculated. The experimental results show that the average accuracy in the peak state of the fast and intelligent measurement method of concrete aggregate volume based on monocular vision is 93.06%, the average accuracy of the volume in the valley state is 92.20%, and the real-time monitoring speed can reach between 111 and 115f/s. The effects can meet the real-time measurement of aggregate volume in the high-level storage bin.