1

I'll admit that the original version of this question was confusing. It "conflated" two issues, watermarking and beaconing. There's no question about why the original was closed.

I removed one of the paragraphs (on "beaconing") and limited the scope of the question to "watermarking." I'm having trouble understanding why it remains closed,

  1. Is the question acceptable for the site in its most recent form? If no, why not?

  2. If yes, should I ask it as a "new" question?

  3. Or should I go through the re-opening procedure, since it is an "old" question.

Below is a "clean" version of the most recent question:

What types of security are used for watermarking applications?

One use of "watermarking" is what I call "branding." That is, you put "watermarks" on your documents to identify them as yours, before sending them out over the "net." These watermarks are usually invisible, but perhaps they need not be, if all they do is to act as your "signature."

One danger of watermarking is "tampering." Another is "altering." What are standard or desirable security measures for these problems?

1 Answer 1

Reset to default
4

On the reopening status, I can't see why the previous question has been closed, but let me guess on "too broad".

I think the newest question might still be broad. Watermarking encompasses a lot of concepts, some of them having ties to security.

You might want to search a bit more on the subject to refine what you are having as problem. Let me give you some clues on the vocabulary:

Watermarking can be robust or fragile. Robust watermarking are meant to resist tampering (or altering, it's the same thing to me). This means that the mark would be inserted in a way that modifications on the media does not destroy it, or that the mark can be recovered using redundancy for example. If the watermark is fragile, this means that any modification will make the mark disappear. The robust-fragile property is a scale and you can find watermark that are semi-fragile (resisting different modifications) or XX-robust (jpeg-robust, ...).

Watermarking can be visible or invisible. The former being a "TV"-logo image on the bottom-left of the screen (for example) and the latter being a hidden code.

And more properties... (blind, informed, ...)

All these properties usually are tied together as making a mark robust will usually translate in stronger modification on the image, thus making the mark visible.

The security feature of watermarking depends on the application. For example, a DRM-protection will use a robust watermarking scheme, visible or invisible, but a watermarking meant to serve as proof by assessing non modification of an image will be fragile (any modification voids the proof by destroying the mark).

Anyway, I suggest you:

  • identify your application first (what are you trying to use watermarking for, is it DRM, identification or authentication)
  • define your field of operation (what kind of media, what kind of message)
  • what do you want to be protected against.

From there, there are a lot of things you can do with watermark. Most of the techniques can be found within the image processing community. Since I have this at hand's reach, let me share an extract of my bibliography (I do not guarantee full access to all these docs):

Watermarking security part one : Theory and Practice. (François Cayre, Caroline Fontaine et Teddy Furon)

Digital watermarking and steganography. (Ingemar J. Cox, Matthew L. Miller, Jeffrey A. Bloom, Jessica J. Fridrich et Ton Kalker.)

A survey of watermarking security. (Teddy Furon)

0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .