Fractal Decoded Image Quality Prediction Based on Accumulated Collage Error Coefficient

摘要: To predict the fractal decoded image quality more efficiently, an accumulated collage error coefficient (ACEC) based method was proposed in this study. Firstly, the definition of ACEC was introduced to describe the relationship among the upper bound, lower bound, and actual value of the accumulated collage error of all range blocks. Moreover, the definition and monotonicity of the relative error of ACEC were also defined and discussed. While the relative error of ACEC reaches a relatively small value, the average collage error (ACER) can be estimated approximately, and then the encoding process can be terminated to directly predict the peak signal-to-noise ratio (PSNR) quality of decoded images. Experimental results show that compared with the previous method, the proposed method can predict the decoded image quality more accurately with fewer computations.