A high-precision measurement method for driver height of shared cars based on MediaPipe

Abstract Aiming at the current problem that it is difficult to automatically adjust shared car seats according to drivers with different heights, a high-precision measurement method for driver height of shared cars based on MediaPipe applied to the shared car automatic seat adjustment system is proposed. First, the monocular camera is calibrated using Zhang Zhengyou calibration method to obtain accurate intrinsic and extrinsic parameters of the camera. Then, the calibrated camera is used to capture the driver’s images. Further, the images are processed by the processing terminal, which detects the key points of the human body in the image utilising MediaPipe and extracts the coordinates of the key points in order to calculate the pixel height of the driver. Finally, the distance between the driver and the camera is measured by the range sensor, and the height of the driver is preliminarily calculated by combining the geometric principle of camera imaging, and the height correction function is introduced to correct the calculation result to obtain a more accurate height value. After experimental verification, the height measurement method proposed in this paper can make the relative error of height control within 0.5%, demonstrating high-precision, and is capable of meeting the demands of the shared car seat automatic adjustment system to automatically adjust the car seat according to the driver’s height.