Image compression method and system based on deep learning

The invention discloses an image compression method and system based on deep learning. A target image to be deeply compressed is acquired; encoding the target image by using an encoder of the image compression model to obtain an encoding file of the target image; decomposing the coding file into a plurality of data streams, and sending the data streams to a server through a Tiantong satellite; and the server receives the coding file and decodes the coding file through a decoder of the image compression model to obtain a restored image of the target image. The image compression model used by the image compression method provided by the invention not only can pay attention to the local features of the image, but also can pay attention to the global features of the image, so that the deep compression image can be better decoded and restored, the fuzzy distortion degree of the decoded image is reduced, and the user experience is improved. 本申请公开了一种基于深度学习的图像压缩方法及系统。通过获取待深度压缩的目标图像；利用图像压缩模型的编码器对目标图像进行编码，得到目标图像的编码文件；将