Abstract: We propose WaveOptima algorithm based on an improved WaveFront (WF) algorithm to address issues such as excessive path turns, high repetitive coverage rate, and low planning efficiency in 3D underwater dam detection. The algorithm employs voxel surface adjacency line-of-sight projection technology to extract surface viewpoints and optimize the viewpoint layout, reducing approximately 40% of redundant viewpoints. It adopts a 26-neighborhood search strategy to reduce both the number of waypoints and the total path length, thereby improving path smoothness and planning efficiency. The algorithm introduces directional and planar costs into the cost function to guide the underwater robot's movement direction and reduce the energy consumption, while also avoiding local optima through a preset direction vector and resolving dead-end problems in complex structures by combining a dynamic escape strategy with a motion priority strategy. Experimental results show that, compared with the traditional WF algorithm and its improved variant, the WaveOptima algorithm reduces the number of waypoints, overlap coverage ratio, and planning time by over 40%, 15%, and 38%, respectively, while significantly improving path smoothness and planning efficiency, making it suitable for various dam structures.