With the advancement of ocean exploration, Autonomous Underwater Vehicles (AUVs) have become essential for marine detection. Equipped with side-scan sonar (SSS), AUVs collect vast amounts of marine environment data. However, the substantial data volume and bandwidth limitations in underwater communication often prevent real-time transmission of SSS images to the mother ship, causing delays in data analysis. To addresses this problem, this paper proposes a SSS image compression and reconstruction algorithm based on Singular Value Decomposition (SVD). Firstly, the SSS image is decomposed into an SVD matrix, and compressed using the first K singular values. Secondly, SSS image restoration is achieved through SVD inverse transformation. By utilizing SVD algorithm, the SSS images can be compressed while ensuring the visual quality of the reconstructed images, thereby promoting cross-domain data transmission between underwater and surfaces. The effectiveness of the proposed method has been validated through extensive real-world experiments.