Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It only takes a minute to sign up.
Sign up to join this community
I am working in digital image processing field, recently I am studying about Image Noise, I just want to know that whether this noise removal is part of image enhancement or image restoration.
I have read some papers on it, at some points authors are saying that image enhancement from a given noisy image, and on the other hand in some papers authors are saying that noise removal is image restoration process.
Please explain what is correct?.
Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.
If you want work in the image processing field, my first advice would be to try and understand various techniques, tools and algorithms used, understand how they work, where and why they are applied and what you can achieve by applying each of them to your input without actually trying to classify them into labeled boxes.
Unfortunately, terminology is not perfect and it is hard to reach a naming consensus even concerning high level concepts, let alone detailed techniques.
But, about noise and noise removal: First, it is important to understand that noise can appear in the image for various reasons most of which could be classified to:
And I would say that noise removal could be part of both processes of image enhancement and image restoration. The opinions are based on my own intuition as somebody who's been working in image processing for a few years now, but:
Image enhancement
If I got an image from a bad-quality sensor, or simply wanted to increase the details on any image obtained directly from a camera, I would call it image enhancement.
An example I saw and would probably classify as such is processing of satellite images. Usually, before processing the image, you want to enhance the contours and remove small intensity variations.
Image restoration
If something corrupted the image you got from the sensor and you want to revert back and obtain an image as close to the original, sensor-returned image, I would call it image restoration.
An example could maybe be trying to correct the images (frames) while video streaming, where you would want to handle the losses as much as possible and still display a (good) video.
It depends on the final goal and the model used.
If we assume we have an image - I which is a result of a sum on of an image W with noise n: I = W + n and our goal by reducing the noise is estimating W than this process is restoration.
Yet I could try achieving a different goal by denoising, such as making it look better (Subjective target) or use prior such as Piece Wise Constant image which means the denoising process if more of an enhancement to the image.
