Accurate measurement of helicopter rotor blade flapping, and analysis of its characteristic patterns are essential for evaluating rotor aerodynamic performance and optimizing structural design. To improve flapping visual measurement accuracy and enable full-phase measurement, this study develops a rotating stereo vision measurement system, proposes a corresponding flapping measurement method, and conducts analysis on blade flapping patterns and laws. First, via photoelectric slip rings, the 10-gigabit-level image data transmission from the rotor end to the ground end is realized. Meanwhile, a symmetric camera mounting bracket is designed to ensure the overall dynamic balance between the cameras and the rotor. Additionally, full-phase blade images are acquired on a rotor test rig under different rotational speeds, collective pitches, and cyclic pitches. Second, a small-target detection network is developed to handle complex illumination inference, enabling high-precision localization of the central pixel coordinates of tiny self-luminous marker points and improving the accuracy of their 3D coordinate computation. Third, a hub coordinate system is established for each acquisition phase. Marker 3D coordinates are transformed from camera to hub system to calculate phase-specific flapping, reducing errors from coordinate drift caused by camera-rotor rotation. Finally, third-order polynomial fitting analyzes in-phase flapping spatial patterns, while composite sine fitting analyzes flapping time-domain laws in the rotation cycle, supporting rotor system optimization. Experimental results on the rotor test rig demonstrate that, within a 1.5 m×1.5 m field of view, static and dynamic flapping measurement errors are 0.44 and 0.82 mm, respectively, both flapping patterns and laws models exhibit excellent agreement with experimental data (RMSE<1 mm). These results verify the effectiveness and high-precision characteristics of the proposed measurement system and method, and this system has been applied to the verification of rotor design tests.