School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China

---

Abstract—This paper presents an auto-recognition method for pointer-type meter based on computer vision. Two calibrated cameras are located in the right and left sides of the meter to capture the meter displayed images. In each displayed image, the fast Hough transform method is used to locate the approximate pointer area and then the least squares fitting method is used to determine the precise line that represents the pointer indicator. Two precise lines obtained from both right and left displayed images are reconstructed in three-dimensional space according to the epipolar constraint. The line reconstructed is then projected into the target plane to determine the true indication of the meter. Experiments show that the proposed method works very well for the dial pointer identification. The maximum uncertainty in the determination of the pointer’s indication is less than the human eye can discriminate.

Index Terms—binocular vision, three-dimensional reconstruction, fast Hough transform, least squares fitting

[PDF]

Cite: Biao Yang, Guoyu Lin and Weigong Zhang, "Auto-recognition Method for Pointer-type Meter Based on Binocular Vision," Journal of Computers vol. 9, no. 4, pp. 787-793, 2014.