Amplitude-modulated continuous wave (AMCW) time-of-flight (ToF) sensor is widely used to capture 3D information of objects due to its relatively high measurement precision in short range. However, the measurement accuracy of AMCW ToF measurement method is generally sensitive to the reflectivity of object, internal stray light, modulation instability, and external light. Consequently, distance measurement error inevitably occurs even in indoor measurement condition. To compensate such error, a post processing method based on machine learning is proposed in this paper. This data driven correction method is validated under indoor measurement condition. According to the experimental results, the distance measurement error correction method presented in this paper shows the most high accuracy compared to other related research results.
|