A sea surface motion field inversion method based on sea surface temperature time series images
By constructing masks, calculating weights and time derivatives, the problems of large-scale background interference and uneven time intervals in sea surface temperature images are solved, achieving high-precision and stable sea surface motion field inversion. It is particularly suitable for complex nearshore and strait areas and can output a variety of business results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF INFORMATION SCI & TECH
- Filing Date
- 2026-05-15
- Publication Date
- 2026-07-21
AI Technical Summary
Existing methods for inverting sea surface motion fields based on sea surface temperature time-series images suffer from problems such as large-scale background gradient interference, difficulty in tracking weak texture regions, distortion of time derivatives due to uneven time intervals, impact of edges and invalid regions on inversion stability, and lack of quantitative zoning analysis capabilities for nearshore waters.
By constructing effective seawater masks, land masks, and other coded masks, high-pass residual images are obtained and texture weights and edge attenuation weights are calculated. The center difference time derivative of three frames is combined with the real time span. The motion vector field is obtained using a dense motion estimation model, and quality evaluation and feedback adjustment are performed. Finally, nearshore and offshore areas are divided for velocity statistics.
It improves the accuracy and stability of sea surface motion field inversion, adapts to complex time-series conditions, outputs velocity field images and quantitative statistical tables, and is applicable to nearshore and strait areas with severe cloud obscuration and complex land-sea boundaries, thus enhancing the interpretability and applicability of the results.
Smart Images

Figure CN122200416B_ABST