Author(s)

Mingxiu Zhang, Jindong Zhang, Mingzhu Zhu, Wenda Liu, Jiatong Tu

Corresponding Author

Jindong Zhang

ABSTRACT

In order to improve the efficiency and accuracy of manual detection of automotive dashboard testing, it is particularly important to use computer vision-related technology to identify pointer readings and icon information of automobile dashboard in automotive dashboard function detection. In this paper, the traditional computer vision technology is applied to identify the automobile dashboard. The pointer reading is recognized by the Hough line detection, and the template matching is applied in each ROI area to determine the lighting and extinguishing of the dashboard indicator. The method realizes the reading of the tread gauge, the speedometer and other pointer instruments of the auto dashboard, as well as the lighting and extinguishing of the steering lights, fog lights and other automotive indicators. The experimental results show that the accuracy of the algorithm's recognition of pointer readings on the dashboard is more than 90%, the accuracy of the identification of indicators is more than 75%, and the average time of frame processing is currently 4.516s, which still needs to be improved.

KEYWORDS

Feature extraction, image processing, automotive instrument clusters, reading recognition, ROI area, template matching