Develop Reference

Compressing images, is there a point?

I'm working on a web site which will host thousands of user uploaded images in the formats, .png, .jpeg, and .gif.
Since there will be such a huge amount of images, saving just a few kb of space per file will in the end mean quite a lot on total storage requirements.
My first thought was to enable windows folder compression on the folder that the files are stored in (using a Windows / IIS server). On a total of 1Gb of data the total space saved on this was ~200kb.
This to me seems like a poor result. I therefore went to check if the windows folder compression could be tweaked but according to this post it cant be: NTFS compressed folders
My next though was then that I could use libraries such as Seven Zip Sharp to compress the files individually as I save them. But before I did this I went to test a few different compression programs on a few images.
The results on a 7Mb .gif was that
7z, Compress to .z7 = 1kb space saved
7z, Compress to .zip = 2kb space INCREASE
windows, native zip = 4kb space saved.
So this leaves me with two thoughs.. the zipping programs I'm using aren't very good, or images are pretty much already compressed as far as they can be (..and I'm surprised that windows built in compression is better than 7z).
So my question is, is there any way to decrease the filesize of an image archive consisting of the image formats listed above?

the zipping programs I'm using suck, or images are pretty much already compressed as far as they can be
Most common image formats are already compressed (PNG, JPEG, etc). Compressing a file twice will almost never yield any positive result, most likely it will only increase the file size.
So my question is, is there any way to decrease the filesize of an image archive consisting of the image formats listed above?
No, not likely. Compressed files might have at most a little more to give, but you have specialize on images itself, not the compression algoritm. Some good options are available in the post of Robert Levy. A tool I used to strip out metadata is PNGOUT.

Most users will likely be uploading files that have a basic level of compression already done on them so that's why you aren't seeing a ton of benefit. Some users may be uploading uncompressed files though in which case your attempts would make a difference.
That said, image compression should be thought of as a unique field from normal file compression. Normal file compression techniques will be "lossless", ensuring that every bit of the file is restored when the file is uncompressed - images (and other media) can be compressed in "lossy" ways without degrading the file to an unacceptable level.
There are specialized tools such which you can use to do things like strip out metadata, apply a slight blur, perform sampling, reduce quality, reduce dimensions, etc. Have a look at the answer here for a good example: Recommendation for compressing JPG files with ImageMagick. The top answer took the example file from 264kb to 170kb.

Remove all metadata from common image formats?

I'm writing a service for a project that's going to handle our image processing. One such process is supposed to strip all metadata from the byte[] provided and return the same image as a byte[].
The method I'm currently working on involves always converting the image to a Bitmap, then converting it back to the original format and returning the data from a MemoryStream.
I haven't been able to test it yet but something tells me I'm going to experience some quality loss.
How can I remove all metadata from any image with a common format?
(bmp, gif, png, jpg, icon, tiff)
Not sure how I can narrow that down any further. Would be nice if I got some feedback regarding the downvotes.

For the lossless formats (except JPEG), your idea of loading it as a bitmap and re-saving is fine. Not sure if .NET natively supports TIFFs (I doubt it does).
For JPEGs, as you suggested there may be quality loss if you're re-compressing the file after decompressing it. For that, you might try the ExifLibrary and see if that has anything. If not, there are command line tools (like ImageMagick) that can strip metadata. (If you use ImageMagick, you're all set, since it supports all of your required formats. The command you want is convert -strip.)
For TIFFs, .NET has built-in TiffBitmapDecoder and ...Encoder classes you might be able to use; see here.
In short, using an external tool like ImageMagick is definitely the easiest solution. If you can't use an external tool, you're almost certainly going to need to special-case the formats that .NET doesn't support natively (and the lossy JPEG).
EDIT: I just read that ImageMagick doesn't do lossless stripping with JPEGs, sorry. I guess using the library I linked above, or some other JPEG library, is the best I can think of.

is there a way to compress a paged .tiff file using C#?

At the end of my process, I need to upload several paged .tiff file images to a website. The files need to be very small, 500kb or less when i upload them.
The problem is, even with me resizing them a lot but at the same time being able to read a few lines of text that are in some of them, they are around 1mb each or so.
I first resize all images going into the tiff files but it's not enough. I need a way to change the quality of them to decrease their size as well.
Can C# do this or would I need a third party software to do it?
The files being uploaded MUST be .tiff.

You don't provide much detail about your data, so can only make some guesses as to what you might need to look at.
First, can you loose some resolution? Can you make the images smaller?
Second, can you loose some color depth? Are you saving the files in a color format when bilevel or greyscale images would suffice?
Third, how clean are these images? Are they photos, scanned documents, what? If they are scanned documents of text or drawings, then some pre-processing to remove noise can make a significant difference in size.
Lastly, what compression method are you saving the file with? Only a lossy format is going to give you the highest degree of compression is most circumstances.
Based on your follow-up:
1) If you can make smaller, this of course saves significant storage space. Determine what is the minimum acceptable resolution that they need to be and standardize on that.
2) If you need to persist color, then this step might not be as effective, since you would have to algorithmically decrease the dynamic range of colors used in the image to an acceptable level before compressing. If you are not sure what this means, then you would probably best skip considering this completely unless you can spend time learning more about image processing and/or using a image processing library that will simplify this for you.
3) I don't think you addressed this in your comments. If you want more precise help, you should update your original question and add much more detail about what you are trying to accomplish. Provide some explanations of what/why you need to do in order to help determine what tradeoffs make sense.
4) Yes, JPG is a lossy format, but I think you may be confusing a few different things (or I may not be understanding your intent from your description). If you are first resizing your original images down into a new JPG file (an intermediate image file), then you are building a TIFF file and inserting the resized JPG as a source image into a multi-page TIFF and saving that, then you need to realize that the process of how the files are compressed in the intermediate files do not necessarily have any correlation with the compression format used in the TIFF file. Depending on what you are using to build and create the TIFF file, the compression format used in the TIFF is done separately and you probably need to specify those parameters when you save that file. If this is what you are doing, then the intermediary process of saving the JPG files may be increasing the size a bit.

File extension from System.Drawing.Image

I'm writing a method that needs to save a System.Drawing.Image to a file. Without knowing the original file the Image was created from, is there anyway to determine what file extension it should have?
The best solution I've come up with is to use a Switch/Case statement with the value of Image.RawFormat.
Does it even matter that I save the Image in it's original format? Is an Image generated from a PNG any different from say one generated from a JPEG? Or is the data stored in an Image object completely generic?

While Steve Danner is correct in that an image created from a JPG will look different to an image created from a PNG once it's loaded into memory it's an uncompressed data stream.
This means that you can save it out to any file format you want.
However, if you load a JPG image and then save it as another JPG you are throwing away more information due to the compression algorithm. If you do this repeatedly you will eventually lose the image.
If you can I'd recommend always saving as PNG.

Image.RawFormat has cooties, stay away from it. I've seen several reports of it having no legal value for no apparent reason. Undiagnosed as yet.
You are quite right, it doesn't matter what format you save it to. After you loaded the file, the internal format is the same for any bitmap (not vector) with the same pixel format. Generally avoid recompressing jpeg files, they tend to get bigger and acquire more artifacts. Steve mentions multi-frame files, they need to be saved a different way.

Yes, it definitely matters because different fileformats support different features such as compression, multiple frames, etc.
I've always used a switch statement like you have, perhaps baked into an extension method or something.

To answer your question 'Does it even matter that I save the Image in it's original format?' explicitly: Yes, it does, but in a negative way.
When you load the image, it is uncompressed internally to a bitmap (or as ChrisF calls it, an uncompressed data stream). So if the original image used a lossy compression (for example jpeg), saving it in the same format will again result in loss of information (i.e. more artifacts, less detail, lower quality). Especially if you have repeated actions of read - modify - save, this is something to avoid.
(Note that it is also something to avoid if you are not modifying the picture. Just the repeated decompress - compress cycles will degrade the image quality).
So if disk space is not an issue here (and it usually isn't in the age of hard disks that are big enough for HD video), always store any intermediate pictures in lossless compression formats, or uncompressed. You may consider saving the finall output in a compressed format, depending on what you use it for. (If you want to present those final pictures on the web, jpeg or png would be good choices).

Convert multiple byte[] to multipage tiff equivalent of Image.SaveAdd without GDI+

I am working on a system that stores many images in a database as byte[]. Each byte[] is a multi page tiff already, but I have a need to retrieve the images, converting them all to one multi page tiff. The system was previously using the System.Drawing.Image classes, with Save and SaveAdd - this was nice in that it saves the file progressively, and therefore memory usage was minimal, however GDI+ concurrency issues were encountered - this is running on a back end in COM+.
The methods were converted to use the System.Windows.Media.Imaging classes, TiffBitmapDecoder and TiffBitmapEncoder, with a bit of massaging in between. This resolved the concurrency issue, but I am struggling to find a way to save the image progressively (i.e. frame by frame) to limit memory usage, and therefore the size of images that can be manipulated is much lower (i.e. I created a test 1.2GB image using the GDI+ classes, and could have gone on, but could only create a ~600MB file using the other method).
Is there any way to progressively save a multi page tiff image to avoid memory issues? If Save is called on the TiffBitmapEncoder more than once an error is thrown.

I think I would use the standard .NET way to decode the tiff images and write my own tiff encoder that can write progressively to disk. The tiff format specifications are public.
Decoding a tiff is not that easy, that's why I would use the TiffBitmapDecoder for this. Encoding is is easier, so I think it is doable to write an encoder that you can feed with separate frames and that is writing the necessary data progressively to disk. You'll probably have to update the header of the resulting tiff once you are ready to update the IFD (Image File Directory) entry.
Good luck!

I've done this via LibTIFF.NET I can handle multi-gigabyte images this way with no pain. See my question at
Using LibTIFF from c# to access tiled tiff images
Although I use it for tiled access, the memory issues are similar. LibTIFF allows full access to all TIFF functions, so files can be read and stored in a directory-like manner.
The other thing worth noting is the differences between GDI on different windows versions. See GDI .NET exceptions.