IMAGE RESTORATION AND IMAGE DECOMPOSTION
 | Image Decomposition |
| Denoising |
| Deblurring |
| Joint Denoising and Deconvolution |
IMAGE DECOMPOSITION
NOISE REMOVAL
Nonlinear total variation (TV) based noise removal algorithm was introduced by Leonid Rudin, Stanley Osher and Emad Fatemi. The total variation of the image is minimized subject to constraints. Here, the algorithm is used to denoise a variety of images.
| Original Image | Noisy Image | Denoised Image |
 |
 |
 |
 |
|
 |
|
|
|
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
Movies: |
 |
 |
|
An example to the right shows a picture taken using a personal digital camera with a desire to get a clear photo. The total variation based noise removal algorithm is used to denoise the image. |
 |
DEBLURRING
Total variation-based image restoration can be used to enhance blurry images. In the examples below, it is seen that a lot of information can be recovered even from extremely blurry images.
| Original Image | Blurred Image | Recovered Image |
 |
 |
 |
 |
 |
 |
DENOISING AND DEBLURRING
Image restoration becomes even harder to accomplish if an image was damaged with both blur and noise. Such images could be improved considerably using total variation image restoration.
| Original Image | Blurred and Noisy Image | Recovered Image |
|
|
 |
 |
|
|
 |
 |
