基于帧内压缩的三维图像声呐自适应压缩方法OA

To alleviate the pressure of real-time transmission of image data to the host computer via network proto-cols,an image compression method based on intra-frame compression is proposed.Considering the characteristics of sonar images,the proposed method simplifies the compression process to the luminance channel of 2D slices ex-tracted from 3D volumetric data.Furthermore,by incorporating Human Visual System characteristics,an adaptive quantization scheme based on block activity analysis is implemented.Experimental results indicate that the pro-posed algorithm achieves a compression ratio of 21.3 while maintaining a Peak Signal-to-Noise Ratio of 34.6 dB and a structural imilarity index of 0.91.With a dual-parallel configuration,thereal-time processing rate reaches 30.6 kHz,and the maximum required network bandwidth is 401.1 Mbit/s.While ensuring image quality and real-time performance,the proposed method successfully resolves the transmission bottleneck caused by the massive data volume of 3D sonar systems and limited network bandwidth,providing effective technical support for the real-time application of underwater 3D imaging systems.