Based on the digital surface model (DSM) and jump point search (JPS) algorithm, this study proposed a novel approach to detect the optimal seamline for orthoimage mosaicking. By threshold segmentation, DSM was first identified as ground regions and obstacle regions (e.g., buildings, trees, and cars). Then, the mathematical morphology method was used to make the edge of obstacles more prominent. Subsequently, the processed DSM was considered as a uniform-cost grid map, and the JPS algorithm was improved and employed to search for key jump points in the map. Meanwhile, the jump points would be evaluated according to an optimized function, finally generating a minimum cost path as the optimal seamline. Furthermore, the search strategy was modified to avoid search failure when the search map was completely blocked by obstacles in the search direction. Comparison of the proposed method and the Dijkstra’s algorithm was carried out based on two groups of image data with different characteristics. Results showed the following: (1) the proposed method could detect better seamlines near the centerlines of the overlap regions, crossing far fewer ground objects; (2) the efficiency and resource consumption were greatly improved since the improved JPS algorithm skips many image pixels without them being explicitly evaluated. In general, based on DSM, the proposed method combining threshold segmentation, mathematical morphology, and improved JPS algorithms was helpful for detecting the optimal seamline for orthoimage mosaicking.

This study proposes a hybrid sea level prediction model by coupling a dynamically optimized variational mode decomposition (VMD) with a convolutional bidirectional gated recurrent unit (CNN-BiGRU). The VMD decomposition is fine-tuned using the grey wolf optimizer and evaluated via entropy criteria to minimize mode mixing. The resulting components are processed by CNN-BiGRU to capture spatial features and temporal dependencies, and predictions are reconstructed from the integrated outputs. Validated on monthly sea level data from Kanmen and Zhapo stations, the model achieves high accuracy with an RMSE of 13.857 mm and 16.230 mm, MAE of 10.659 mm and 13.129 mm, and NSE of 0.986 and 0.980. With a 6-month time step, the proposed strategy effectively captures both periodic and trend signals, demonstrating strong dynamic tracking and error convergence. It significantly improves prediction accuracy and provides reliable support for storm surge warning and coastal management.