峰值信號雜訊比 (Peak Signal to Noise Ratio)
通常在經過影像壓縮之後,輸出的影像通常都會有某種程度與原始影像不一樣。為了衡量經過處理後的影像品質,我們通常會參考PSNR值(峰值信號雜訊比 (Peak Signal to Noise Ratio))來認定某個處理程序是否令人滿意。來看Lena如何決定PSNR吧~

Denoising results using different methods in the DWT domain. (a) Clean image; (b) noisy image: PSNR=22.16 dB (σ=20); (c) denoised image using BiSrhink: PSNR=31.09 dB; (d) denoised image using ProbShrink: PSNR=31.00 dB; (e) denoised image using SURE-LET: PSNR=31.02 dB and (f) denoised image using the proposed method: PSNR=31.26 dB.
![[OpenCV] 影像拼接—寬景和環景 (Image Stitching - Wide View and Round View)](https://cg2010studio.com/wp-content/uploads/2012/05/opencv-e5bdb1e5838fe68bbce68ea5e28094e5aface699afe5928ce792b0e699af-image-stitching-wide-view-and-round-view.jpg?w=540&h=239)





HappyMan・迴響