A real-time compressing method for complex SAR images

Complex image compression is widely used in diverse remote sensing applications in order to reduce the costs of data storage and release the burden of data transmission on limited data transmission channels. However, the huge amount of data makes it difficult to compress in real-time. A novel real-time compressing method for complex SAR images is presented in this paper. The system uses a fixed-point DSP of TI SM320c6416 with IGB SDRAM (synchronous dynamic random access memory). After removing the redundancy of complex image in frequency domain, arithmetic codec is applied for statistical redundancy, further. With the least data regress limes, good performance is achieved. The results show that compressing a complex image of 8192*2048 needs only 4315 ms in an acceptable distortion with compression ratio of 16.214:1. So, the method is highly suitable for real-time (or low delay) applications.