Large-Area Marine Surface Current Measurement Using Unmanned Aerial Vehicle-Based Optical Flow Mosaicking Without Ground Control Points

Abstract Nearshore ocean currents are characterized by complex structures and spatial heterogeneity. Conventional measurement techniques are either inefficient or offer low spatial resolution, while image velocimetry (a low-cost, high-resolution method) is limited in marine applications due to the trade-off between large-area coverage and high resolution, as well as the lack of ground control points for calibration. To address these issues, this study proposes a current field mosaicking method. By maintaining the camera in constant-speed motion, the distance differences between consecutive image frames are converted into row offsets using the interframe time interval and image resolution, thereby expanding the velocimetry observation range through stitching. An adaptive thresholding method combining sun-glint identification and cropping is also developed. Results show that the picture range increases from 60 to 350 m, enabling large-area optical-flow velocimetry, and that the sun glint removal technique effectively reduces interference without additional quality degradation or artificial signals. This approach solves key constraints of image velocimetry in large-scale marine environments.