Traditional fractal image decoding is very fast, and the process is difficult to control. Special applications such as video effects and animation may require a gradual image decoding process. Some scholars have introduced control parameters into compression mapping to slow down the fractal decoding process, but the relationship between control parameters and the peak signal-to-noise ratio (PSNR) of decoded image sequences has not been established. In this paper, a fractal image decoding algorithm based on peak signal-to-noise ratio is proposed, which realizes that the decoding process can be conducted according to the designed peak signal-to-noise ratio sequence, providing options for diverse image decoding applications.