Making Do With Less Data

If you wanted to make an image you could simply omit or average every other pixel.

Is there a short list on the most popular techniques for making do with less data?

Bret Cahill

Are we conserving data now Bret? How about omitting every other dollar from your pay? or, Omitting other from posts?

:-) Tom

JPEG

John

Take a long list of such techniques and then apply one of them...

-- When a true genius appears in the world, you may know him by this sign, that the dunces are all in confederacy against him. Jonathan Swift: Thoughts on Various Subjects, Moral and Diverting

PNG

quasi

Probably none of the groups sci.math.num-analysis, sci.math, or sci.electronics.basics are the best places for your question; perhaps try or sci.image.processing, or google for image processing forums.

You haven't said with any precision what it is you want to accomplish. Some other posters mention jpeg and png image storage methods, which apply data compression methods to images, but my impression is that what you are asking about is methods of image resizing.

As a starting point, see , which lists several resizing methods: Bicubic interpolation, Bilinear interpolation, Lanczos resampling (windowed sinc filter), Seam carving (as in ) Spline interpolation, and Supersampling, but for some reason doesn't mention Genuine Fractals (see following URL for that).

For some brief explanations about resizing and some example pictures:

Image processing programs like gimp, Photoshop, and Paint each implement several of the methods mentioned above.

TIFF format:

GIF format

In my opinion, PNG format is the best overall.

quasi

Ok, you win, your list is shorter than mine.

John

True but this is a very general discussion.

A year or so ago NPR had something about saving time [$] on CAT scans or MRIs.

Bret Cahill

The way I heard it, JPEG is a "lossy" compression algorithm, and the others aren't, for whatever that's worth. Then again, Bret was talking about throwing away data in the first place.

Thanks, Rich

How about omitting every other Cahill post? ...or better, all of them.

Why not just use his posts as a source of cheap amusement, like we do with yours and Jim's and Michael's? >:->

Starbuck's kind of fun sometimes, and Phil's always good for at least a groan. >:->

One of the biggest laughs I got was at a seminar, where the lecturer was talking about how to deal with stupid traffic without giving yourself apoplexy - I got the mic, and said, "Think of the antics of the stupid drivers as a form of entertainment."

Laughter ensued. It felt good. ;-)

Cheers! Rich

In a sense, this idea is spot on. Certain image compression techniques will omit the spatial high frequency components in an attempt to get by with much less data. This involves "looking" at the image in a different way, i.e. its frequency space via DFT (FFT).

_Fewer_ data.

-- When a true genius appears in the world, you may know him by this sign, that the dunces are all in confederacy against him. Jonathan Swift: Thoughts on Various Subjects, Moral and Diverting

Frederick Williams wrote in news: snipped-for-privacy@btinternet.com:

No. "Less" is correct. Data is both plural and singular in construction, so it may correctly take singular verbs, singular modifiers and singular pronouns. English is not Latin; "data" is not just the plural of "datum."

There are two ways to make a pile of data smaller:

1. Lossless compression
2. Lossy compression Try nosing around over in news:comp.compression after reading their FAQ

Since the original question concerns "making do with less data," your response is not strictly correct.

Lossless compression discards nothing, hence we are not not making do with less.

Lossy compression does discard data, although not in the spatial intensity domain that the OP assumes. Thus only by using lossy compression do we make do with less.

Here's an idea... First figure out how many different colors the image has. Store just one pixel of each color in the file (huge savings right there!) Then map the position of each color in x, y coordinates.

When loaded the mostly text file simply copies a color pixel and puts it in all the locations specified in the coordinates.