Images come under different kinds of attacks when sent over a transmission medium. They may be corrupted with noise or the image data may undergo compression at sender or elsewhere.
What are the other possible attacks that may happen just before or during image data transmission irrespective of what content the image is carrying.
Blur or the wather (such as fog, rain) may also attack image quality during acquisition process.
Before is guass.noise and after depend on the environment of compress and transmission
In data hiding applications there are some intentional attacks as for example:
collision atack, repudiation attack, ...
See chapter 2 of the following PhD Thesis:
http://www.tc.etc.upt.ro/personal/corina/book05.pdf
I would recommend that you categorise the types of degradation.
The largest "deviation" is a misplaced object, due to an indexing error of the file! Let us not forget this case. Call it extra-image error.
Send me a picture of Rome and believe it's a picture of Beijing... Semantic error.
Then there is the case of errors within the image, call the intra-image errors. An image if a function F(x,y) where coordinates (x,y) are in a set Z.
You cannot prevent errors from happening. You can mitigate the damage in case of error. Instead of F(x,y), you get F'(x,y)... The goal is that from F'(x,y) "received", you can reconstruct F"(x,y) as close as possible to F(x,y).
The closeness needs to be defined for the particular application you are considering. Do you want to protect the center more? Certain objects in the image more? Then you might change your representation just to do that. Changing a representation is made through a mathematical transformation, which is inverted after "reception".
Sorry, this is a general answer, but your question is general. It leads to a "do it yourself" approach which is likely to work for your problem (but maybe not others).
What matters most is stay in control: use your image properly, understand what you need from it, and adapt the representation and processing.
Dear @Renaud Di Francesco Sir
Your answer is more than what I was looking for or may be an answer to what I was lacking.
Thanks a lot
- Badges
- Science topic
- Similar topics
- Computer Science
- Images