Tracking and measurement of periodic rotating blades after video recombination based on center mark recognition

Based on the computer vision, optical non-contact measurement has been widely applied in the field of mechanical vibration measurement. However, existing computer vision methods are primarily suitable for on-site vibration scenarios, which pose significant technical challenges for combined scenarios involving both rotation and vibration. When measuring high-speed test points, the simplest and most direct approach is to use high-frame-rate image acquisition devices. This paper adopts a gradient-based optical flow tracking and measurement technique to track points of interest on rotating structures. In order to overcome the limitation of optical flow method's “spatial continuity assumption” on the measurement speed of rotating structures, a novel optical tracking and measurement method for periodically rotating blades based on center marker recognition in reorganized videos is proposed. Building upon the optical flow-based target tracking and detection algorithm, an improved method based on Quick Response Code（QR Code） marker tracking and Phase Mapping Method (PMM) is proposed. The optical flow method for rotating structures achieves an increased maximum detection speed, thereby reducing the cost of image sampling devices. In theory, the maximum detectable rotational speed that can be increased depends on the ratio of the length of the blade (R) to the adopted two-dimensional side length (r), denoted as R/r. In this paper, the theoretical maximum value for the increased rotational speed is R/r≈4. Compared to high-frame-rate cameras with costs ranging from tens of thousands to even several hundred thousand dollars, the commercial camera devices used in this paper are priced at around a few thousand dollars. On the other hand, the use of QR Code for marking and tracking improves the accuracy of the detection algorithm and allows for the inclusion of more information within the QR Code. The paper introduces and analyzes the fundamentals and principles of this algorithm, conducts experimental measurements, and verifies the proposed improvement through theoretical calculations.