3D imaging of underwater scanning photon counting lidar based on multiscale spatio-temporal resolution.

This manuscript presents an underwater scanning photon counting lidar system specifically designed for the 3D imaging of underwater targets. A multiscale spatio-temporal resolution method is proposed to enhance the accuracy and resolution of 3D scanning imaging. Signal extraction is achieved through macro pulse accumulation number and macro time resolution, based on the spatio-temporal correlation constrained by relative signal intensity difference factor. Subsequently processing focuses exclusively on the photon-counting events extracted in the previous step. By employing micro pulse accumulation number and micro time resolution, each pixel is expanded into multiple pixels, thereby improving timing precision. This not only enhances imaging accuracy but also enables the detection of more detailed information about the target. The reconstructed images of the resolution plate located approximately 3.5 meters away demonstrate that both the imaging accuracy and resolution are within 10.0 mm. In addition, high-performance 3D reconstruction of a coral model located approximately 4 meters away with complex surface shape is also realized, where each branch of the coral can be distinctly identified. It is verified that the developed lidar system has the ability of high-performance 3D imaging for underwater targets. This lidar system will play an important role in the field of fine mapping of seabed topography and underwater target detection and recognition.