Image Restoration Quality Measurement using Noise Filters
Tushar Debnath, Surajit Paul, Kumar Amitabh · International Journal of Electronics and Communication Engineering · 2023
The aim of this project is to implement different features of image restoration. We consider a vectorised picture and take pictures of it with our mobile for two configurations, the first lit by natural light, the second without the light. Those photos apply different transformations on the original image as rotation, change of the scale, integration of unwanted environment in the picture, etc. We will see a method to restore those photos and compare them to the original image. The comparison will be made by using two indicators: the Peak Signal Noise Ratio (PSNR), measuring the quality of reconstruction of the image, and the Structural Similarity Index (SSIM), evaluating the similarity between pixels. Finally, the complementary analysis will be performed to increase the restoration quality between the pictures, like using a noise reduction filter or function to increase the correspondence between histograms.