Static Camera Calibration for Advanced Driver Assistance System Used in Trucks - Robust Detector of Calibration Points

摘要

The paper presents selected investigation results on a static (factory) calibration procedure of the camera, for the use in Advanced Driver Assistance System (ADAS) mounted in trucks. It is a commercial project that has already been put for production. For this reason, special emphasis was put on the reliability of the proposed solutions. Additionally, the developed algorithms were optimized so that all computations are completed within an assumed period of time (requirements). One of the problems encountered during the development of the calibration procedure was the fact that various objects were located around the calibration pattern in the factory. In some situations these objects may be classified as calibration points, resulting in false outcomes of the overall calibration procedure. For this reason, special emphasis was placed on the development of appropriate robust algorithms responsible for the detection of calibration points. For example, the implemented methods allow for distinguishing the false points and removing them from the list of detected points. The overall procedure was carefully tested under various conditions, including unfavorable ones, with more than 500,000 tests.