Irregular Motion Detection in Automotive Navigation Systems with Soft Map Rotation
Martin Pollot, Dominic Springer, Ralph Schleifer, Monika Nitsch, André Kaup · 2019
Visually unpleasant and erroneous positioning in navigation scenes is an issue for, but not limited to, premium vehicle manufacturers. In order to assure consumers high quality products, intelligent automated display testing is required. This paper presents an enhanced version of an error detection algorithm for navigation sequences based on novelty detection. Therefore, motion parameters obtained from real world navigation sequences are used. These are taken directly from the display using a screen-grabbing device. The proposed algorithm works solely on the information provided in these images. Several image processing methods extract information describing the overall motion between two consecutive frames which is then fed into a novelty detection algorithm to predict outliers. In this case, these outliers describe motions that should not occur in the displayed scene as they display impossible movements of the car. Experimental results demonstrate the improvement over the state of the art, being able to deal with recent changes in the displayed scene in order to hide or conceal present positioning errors. Evaluating the performance of the proposed algorithm, a precision of 76.2%, which is significantly higher than the state of the art method, could be reached while not decreasing the recall at all.