Convert 32bit frame to 24bit fram

Question

2.00/5 (1 vote)

See more:

Hi. I want to convert a 32 bit frame to 24 bit to pass it to AForge filters. I am doing it using graphics as follow:

bitmap bmap = new Bitmap(640, 480 , System.Drawing.Imaging.PixelFormat.Format32bppRgb);
clone = new Bitmap(bmap.Width, bmap.Height, System.Drawing.Imaging.PixelFormat.Format24bppRgb);

using (Graphics gr = Graphics.FromImage(clone))
{
gr.DrawImage(bmap, new System.Drawing.Rectangle(0, 0, clone.Width,
clone.Height));
}

but it takes large amounts of memory, is there any other way to do it?

Thanks in advance

Posted 15-Dec-13 17:48pm

niloufar.M

Add a Solution

Comments

Sergey Alexandrovich Kryukov 16-Dec-13 0:37am

What, too much memory for 640x480? Are you kidding? I hope you want to clone it frame by frame, one at a time. If so, you method is reasonable enough; and the memory consumption should be quite moderate.
—SA

niloufar.M 16-Dec-13 2:07am

Dear SA
Thanks for your reply, my project is real time and I'm capturing from kinect, so it is 30 FPS.

Sergey Alexandrovich Kryukov 16-Dec-13 2:11am

Why would you need to convert on the fly? How about pre-mastering it? Anyway, then your problem could be performance, not memory consumption...
—SA

niloufar.M 16-Dec-13 2:35am

Maybe, I am doing some process which I need it to be real time, so do you have any idea?

Sergey Alexandrovich Kryukov 16-Dec-13 10:10am

Not really. I think your method is good. But, if its performance is insufficient (did you try?), maybe the System.Drawing is not fast enough for your purpose...
—SA

niloufar.M 16-Dec-13 19:41pm

Yes I tried it and it is not fast enough, thanks :)

Sergey Alexandrovich Kryukov 16-Dec-13 20:17pm

Are you saying that AForge.NET operations are not the bottleneck but your simple conversion is? Or all together. Maybe you can skip this step, as you do some double work, together with AForge.NET?
—SA

niloufar.M 17-Dec-13 2:02am

I'm shifting pixels in the bitmap and then I'm trying to convert in to 24bpp, but when I'm doing them together I got problem. I'm shifting pixels but changing the value of each pixel as it is argb to the value o the other pixel I want to shift to it's place. any help?

Sergey Alexandrovich Kryukov 17-Dec-13 3:22am

Yes, maybe... Wait a minute...
—SA

1 solution

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Sergey Alexandrovich Kryukov · Accepted Answer · 2013-12-16T21:38:00

It looks like you have a performance leak. Here is what happens: you to accomplish some task with AForge.NET, but it requires different bit format, so you also convert to the 24-bit pixel format. I think your steps are quite reasonable, but you still get a bit less performance than requited by real-time video. Does it mean that you cannot improve performance (without resorting to native code)? I don't think so.

Here is what happens: you really do at least some of double work. Internally, System.Drawing does a number of things: it reads data, interprets it, put bits to canvas, something like that. And AForge.NET later does at least part of it again. Logically, the resolution would be to remove redundancy. Here are the ideas I have:

Exclude reformatting pixels by looking at the source code of AForge.NET. Modify or update AForge.NET with the methods you use now, but working with your pixel formats. I means considerable job, but I'm sure it's quite possible. You just copy and slightly modify the code.
You say that you just need to shift pixels. It sounds very simple. Most likely, you can work it out without using AForge.NET. You need to lock the pixels (specialized data pinning for pixels available in System.Drawing. Here is how:
http://msdn.microsoft.com/en-us/library/system.drawing.bitmap.lockbits%28v=vs.110%29.aspx[^].

The MSDN articles on two methods referenced above comes with code samples which can help you to understand how to use them. You have two possibilities to improve performance by removing redundant processing:
a) you can have two bitmaps with different pixel format, to change pixel format of the target bitmap, but this time you can do your image transformation in the same "lock bit session", avoiding redundant marshaling of pixel back and forth, which should be a considerable overhear;
b) you can work with the same pixel format, as before, the 32-bit one; in this case, you can do all the transformation on the same set of bits, marshaling the image to data and data to image only once.

Please try. Good luck!

—SA