An Inertial Sensor for Mechanical Compensation of the Vehicle Vertical Movement Impact on In-Vehicle Embedded Camera Orientation

Abstract

Vision systems play an important role in many intelligent transportation Systems (ITS) applications, such as automatic vehicle guidance, driver assistance, traffic law reinforcement and so on. These systems installed in either outdoor environments or in vehicles themselves have often suffered from image instability. In this paper, a low cost inertial sensor devoted to a mechanical compensation system for the vehicle vertical movement impact on an in-vehicle mounted camera is presented. We begin with measuring the accelerations on two different points on the vehicle. These measurements are combined to provide an estimation of the vertical vehicle frame movements. The computed motion consists of both expected and unexpected components. They are discriminated and the expected motion component is further smoothed. The resulting motion is next processed by a specific Kalman filter, aimed to predict the next motion and to provide more reliable motion estimation based on both kinematical model and measurements. The camera orientation is then adjusted such as to meet the viewing requirements. In our case, as was described in [12], [11] the main idea is to preserve the License Plate of the followed vehicle in the field of view of the camera in such manner to minimize the computations to track the License Plate in the images sequences. A series of experimentations with both real and synthetic data have been conducted. The experimental results have revealed the effectiveness of the proposed measurement process.