The present invention concerns a classification technique for evaluating a visibility degradation in image data units (Pi). Image data units (P1, P2, P3, P4) concerning the environment (3) of a vehicle (1, 1') are gathered from a vision sensor (2, 2'). At least a share of the image data units (P1, P2, P3, P4) contains image information which is subject to a visibility degradation based on at least one vision degrading influence (5, 6, 7, 8). The invention suggests various methods and devices for automatically creating filter rules based on an artificial intelligence approach and provides a classification device (4) that can estimate a visibility degradation of newly gathered image data units (P*) by attributing a rank parameter (R*), thereby allowing to judge the reliability of vision sensor data.
評価方法、トレーニングデータセット、エミュレータ作成方法、分類方法、試験方法、車両制御装置、および車両
2021-08-18
Patent
Elektronische Ressource
Japanisch
IPC: | G06T Bilddatenverarbeitung oder Bilddatenerzeugung allgemein , IMAGE DATA PROCESSING OR GENERATION, IN GENERAL / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |