Image Quality Assessment for Measuring the Degradation by Using the Dual-Tree Complex Discrete Wavelet Transform

摘要

We present a new image quality assessment method for measuring the degradation of a digital image. The PSNR and SSIM are widely used to evaluate image quality degradation. However, the results obtained with the PSNR and SSIM do not always match a subjective evaluation. We developed a new image quality assessment method based on the dual-tree complex discrete wavelet transform (DT-CDWT) and a weighted geometric mean for four index values, namely, sharpness, granularity, 2-norm error, and infinity norm error, so as to match a subjective evaluation. We describe our image quality assessment in detail and present experimental results demonstrating the usefulness of our method.