为提高水下目标探测的图像质量,研究一种基于计算机图形的声呐图像生成(Computer Graphics-based Sonar Image Generation,CG-SIG)方法.该方法结合声学物理建模、计算机图形渲染、深度学习优化技术.实验结果表明,CG-SIG方法生成的声呐图像在峰值信噪比(Peak Signal to Noise Ratio,PSNR)、结构相似性(Structural Similarity,SSIM)、平均主观得分(Mean Opinion Score,MOS)指标上均优于传统方法,为水下目标探测提供了可靠的图像支持.
Research on Key Techniques of Sonar Image Generation Based on Computer Graphics
In order to improve the Image quality of underwater target detection,a Computer Graphics-based Sonar Image Generation (CG-SIG) method is studied. The method combines acoustic physical modeling,computer graphics rendering and deep learning optimization techniques. The experimental results show that the sonar image generated by CG-SIG method has the characteristics of Peak Signal to Noise Ratio (PSNR),Structural Similarity (SSIM) and Mean Opinion Score (MOS) are superior to traditional methods,providing reliable image support for underwater target detection.
万方数据
维普
