Update the image source for PSNR page

Author: minhduy13ly

book/6-computer-vision.tex

@@ -91,16 +91,18 @@ \subsection{Peak Signal-to-Noise Ratio}
 
 \textbf{Figure 7.1 Image with different PSNR.} Demonstrates noisy images with their corresponding PSNR values compared to the original ones. \textbf{Left:} Original image and its noisy version. 
 \textbf{Right:} Illustration of how strong noise signals have been added to the original image.
-
-\vspace{-10pt} 
+The color image is from the popular \href{https://data.vision.ee.ethz.ch/cvl/DIV2K/}{DIV2K} dataset.
+\vspace{-12pt} 
 
 \orangebox{Did you know that...}
 {PSNR is not just for images; it can be applied to \textbf{video, audio, or even signal processing}. Anytime we want to measure the quality of \textbf{reconstructed} signal compared to the \textbf{original}, PSNR could be considered.}
 
-
+\vspace{-5pt} 
 
 \textbf{Other related metrics}
 
+\vspace{-5pt} 
+
 Modern computer vision problems often focus on generating likely realistic images that are perceptually pleasing to humans, rather than strictly matching the
 target image pixel by pixel. Therefore, PSNR is typically used in conjunction with other modern metrics like Fréchet Inception Distance (FID) or Inception Score (IS).