校准
计算机视觉
人工智能
计算机科学
计算机图形学(图像)
物理
量子力学
作者
Bin Liu,Yuchen Su,Yihang Li,Guanhao Wu
标识
DOI:10.1109/tim.2025.3588965
摘要
Virtual binocular stereovision system (VBSS) based on biprism provides an effective and compact solution for three-dimensional measurement. The calibration accuracy of the system is important for precise measurement. Generally, the structure parameters need to be optimized before measurement, especially for VBSS due to its narrow field of view, limited depth range and short baseline. In this paper, we proposed a precise calibration method for the structure parameters of VBSS considering its imaging characteristics. The multiple spatial distance constraints were designed and incorporated into the objective function with the epipolar constraint. The structure parameters can be nonlinearly optimized by the improved objective function. Experiments were designed to compare the performance of the proposed method with four existing methods. Furthermore, the optimization strategy of the proposed method was compared with three other strategies. The results demonstrate the feasibility of the proposed method and it can effectively improve the calibration accuracy for VBSS.
科研通智能强力驱动
Strongly Powered by AbleSci AI