High-Definition Sonar Imaging Using 2-D Low-Complexity Adaptive Processing

Forward-looking sonar (FLS) uses matched filtering (MF) and conventional beamforming (CBF) to process the echo and get a 2-D image. The imaging results suffer from low resolution and high sidelobe levels (SLLs), leading to low definition. To solve this problem, we present a 2-D low-complexity adaptive (LCA) sonar imaging method to achieve high-definition images. In the azimuth dimension, we employ a set of predesigned Chebyshev and Kaiser windows, combined with left- and right-steered variations of these windows, to perform angular LCA beamforming. In the range dimension, we use weighted MF with Chebyshev and Kaiser windows to improve the range resolution and reduce range SLLs. In both azimuth and range dimensions, the proposed method adaptively selects the optimal windows from a set of predesigned ones under the constraint of minimum power distortionless response. This approach can be viewed as a discrete form of 2-D adaptive processing, offering improved imaging quality over conventional methods while maintaining robustness and low complexity. Simulation studies are conducted to evaluate the performance of the proposed method. Results show that it outperforms existing sonar imaging methods in key metrics such as half-power beamwidth (HPBW), peak sidelobe level ratio (PSLR), and average sidelobe level (ASL). In addition, the method demonstrates robustness under small array manifold errors, as well as in low signal-to-noise ratio (SNR) and low signal-to-reverberation ratio (SRR) environments. Quantitative image quality assessments based on peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) further confirm the superiority of the proposed method. These improvements suggest that the enhanced imaging performance can be beneficial for underwater target detection and classification tasks. Furthermore, real-data experiments conducted in a lake environment confirm the practical effectiveness of the method in generating high-definition sonar images with enhanced clarity and detail. These findings highlight the practical value of the proposed method in high-definition sonar imaging.