Develop Reference

Conversion from byte[] to image using image.FromStream method

I try converting image data from the database which is already in byte[] back to the image and I'm getting "invalid parameter error" using Image.FileStream.
Please, can anyone help me out with this?
I've tried working around the code using various methods and the last one is below in my code.
byte[] data = validaccount.FingerPrint;
try
{
using (MemoryStream strm = new MemoryStream())
{
strm.Write(data, 0, data.Length);
strm.Position = 0;
System.Drawing.Image img = System.Drawing.Image.FromStream(strm);
BitmapImage bi = new BitmapImage();
bi.BeginInit();
MemoryStream ms = new MemoryStream();
img.Save(ms, System.Drawing.Imaging.ImageFormat.Bmp);
ms.Seek(0, SeekOrigin.Begin);
bi.StreamSource = ms;
bi.EndInit();
}
}
catch (Exception ex)
{
MessageBox.Show(ex.ToString());
}
The code is supposed to convert the byte[] into an image.

According to the docs, Image.FromStream(stream) will throw an argument exception if "The stream does not have a valid image format". Have you verified the data is actually correct? If so, what type of image is it?
You're misusing your streams. You only need 1 memory stream and it has a constructor that takes a byte array (no need to write the bytes yourself). Be sure to wrap it in a using block (like you did for your first stream).
You may not want to use BitmapImage - that's for xaml/wpf apps. You probably want System.Drawing.Bitmap which inherits/extends System.Drawing.Image. Additionally, Bitmap has a constructor which takes a stream - no need to use FromStream.
Finally, Image (and hence Bitmap since Bitmap inherits Image) implements IDisposable, so you should also wrap it in a using block.
P.S. This is a duplicate question.

Though its not memory stream, this method has worked for me and if you browse SO for your question, some times MS does not work.
using System.Drawing;
var converterdImage = (Bitmap)((new ImageConverter()).ConvertFrom(byteArray));
Byte Array to Image Conversion

Destroying stream cause destroying the bitmap

I have something like:
public Bitmap GetBitmap()
{
Byte[] byteArray= bring it from somewhere
using (Stream stream = new MemoryStream(byteArray))
{
return new Bitmap(stream);
}
}
When I use this method outside the Bitmap is crushed. but if I stepped into the "using" scope the bitmap will be exists and works fine. it seems that disposing the stream cause disposing the bitmap..
the question is:
Do I need some deep copy? how should I perform it?

When you dispose the Bitmap will be lost, so indeed you need so perform a deep copy.
Eventually your code should be:
public static Bitmap GetBitmap()
{
byte[] byteArray = bring it from somewhere
using (Stream stream = new MemoryStream(byteArray))
{
var tempBitmap = new Bitmap(stream);
return new Bitmap(tempBitmap); // This will deep-copy the Bitmap
}
}
By the way, usually primitive types, like byte are written in small case.

Difficulty Saving Image To MemoryStream

I'm having some difficulty saving a stream of bytes from an image (in this case, a jpg) to a System.IO.MemoryStream object. The goal is to save the System.Drawing.Image to a MemoryStream, and then use the MemoryStream to write the image to an array of bytes (I ultimately need to insert it into a database). However, inspecting the variable data after the MemoryStream is closed shows that all of the bytes are zero... I'm pretty stumped and not sure where I'm doing wrong...
using (Image image = Image.FromFile(filename))
{
byte[] data;
using (MemoryStream m = new MemoryStream())
{
image.Save(m, image.RawFormat);
data = new byte[m.Length];
m.Write(data, 0, data.Length);
}
// Inspecting data here shows the array to be filled with zeros...
}
Any insights will be much appreciated!

To load data from a stream into an array, you read, not write (and you would need to rewind). But, more simply in this case, ToArray():
using (MemoryStream m = new MemoryStream())
{
image.Save(m, image.RawFormat);
data = m.ToArray();
}

If the purpose is to save the image bytes to a database, you could simply do:
byte[] imgData = System.IO.File.ReadAllBytes(#"path/to/image.extension");
And then plug in your database logic to save the bytes.

I found for another reason this article some seconds ago, maybe you will find it useful:
http://www.codeproject.com/KB/recipes/ImageConverter.aspx
Basically I don't understand why you try to write an empty array over a memory stream that has an image. Is that your way to clean the image?
If that's not the case, read what you have written in your memorystream with ToArray method and assign it to your byte array
And that's all

Try this way, it works for me
MemoryStream ms = new MemoryStream();
Bitmap bmp = new Bitmap(panel1.Width, panel1.Height);
panel1.DrawToBitmap(bmp, panel1.Bounds);
bmp.Save(ms, System.Drawing.Imaging.ImageFormat.Jpeg);
byte[] Pic_arr = new byte[ms.Length];
ms.Position = 0;
ms.Read(Pic_arr, 0, Pic_arr.Length);
ms.Close();
Well instead of a Image control, I used a Panel Control.

Changing Bitmaps using c#

I was working on my college project where i was trying to change bit values of a Bitmap.
I am loading the bitmap to the memory stream, then extracting it to byte[] array. Now I was changing few central bits of this byte[] array and then converting it back to a bitmap image.
But i got a run time exception that "Invalid Bitmap".
Does a bitmap have some special format instead of simple bits???
Following is the code used by me:
MemoryStream mstream = new MemoryStream();
Bitmap b = new Bitmap(#"D:\my_pic.bmp");
b.Save(mstream, System.Drawing.Imaging.ImageFormat.Bmp);
byte[] ba = mstream.ToArray();
mstream.Close();
byte[] _Buffer = null;
System.IO.FileStream _FileStream = new System.IO.FileStream(_FileName, System.IO.FileMode.Open, System.IO.FileAccess.Read);
System.IO.BinaryReader _BinaryReader = new System.IO.BinaryReader(_FileStream);
long _TotalBytes = new System.IO.FileInfo(_FileName).Length;
_Buffer = _BinaryReader.ReadBytes((Int32)_TotalBytes);
// close file reader
_FileStream.Close();
_FileStream.Dispose();
_BinaryReader.Close();
int leng1 = ba.Length;
int leng2=_Buffer.Length;
for(i=(leng1)/2;i<leng1;i++)
{
for(j=0;j<leng2;j++)
{
ba[i]=_Buffer[j];
}
if(j==(leng2-1))
{
break;
}
}
TypeConverter tc = TypeDescriptor.GetConverter(typeof(Bitmap));
Bitmap bitmap1 = (Bitmap)tc.ConvertFrom(ba);

You must have your own goal to want to operate at this low level with a bitmap and that's fine. Unless you are performance bound it is way easier to do graphics at the graphics API level. Even if you are performance sensitive, other people have cut a path through the jungle already.
Back to the question. The BMP file format is simpler than some others but still comes in many varieties. Here is a detailed introduction to the gory details of the BMP format:
BMP file format
Now if you are just parsing your own BMP files and you know they are 32-bit RGB, and the header is going to be such-and-such a size which you can skip over, etc, then that might work for you. If you need to handle any old BMP it gets messy real fast which is why the libraries try to take care of everything for you.

Out of memory error while loading a Bitmap

I'm working with big size images (for ex. 16000x9440 px) and cut some regions for other things. I'm getting an exception "Out of memory" when create a new Bitmap instance:
using (FileStream fileStream = new FileStream(mapFileResized, FileMode.Open))
{
byte[] data = new byte[fileStream.Length];
fileStream.Read(data, 0, data.Length);
using (MemoryStream memoryStream = new MemoryStream(data))
{
using (Bitmap src = new Bitmap(memoryStream)) // <-- exception
{
tile = new Bitmap(tileWidth, tileHeight, PixelFormat.Format24bppRgb);
tile.SetResolution(src.HorizontalResolution, src.VerticalResolution);
tile.MakeTransparent();
using (Graphics grRect = Graphics.FromImage(tile))
{
grRect.CompositingQuality = CompositingQuality.HighQuality;
grRect.SmoothingMode = SmoothingMode.HighQuality;
grRect.DrawImage(
src,
new RectangleF(0, 0, tileWidth, tileHeight),
rTile,
GraphicsUnit.Pixel
);
}
}
}
}
When I use small image sizes (for ex. 8000x4720 px) then all work fine.
How can I work with big size images?
PS tile Bitmap is disposed in finally block.
Best regards, Alex.

You are using about a gigabyte of ram, not very suprising that you run out of memory.
Assuming you use a 32bpp Fileformat with 16000x9440 pixel you get a filesize of about:
16000 * 9440 * (32/8) = ~576MB
byte[] data = new byte[fileStream.Length];
fileStream.Read(data, 0, data.Length);
using (MemoryStream memoryStream = new MemoryStream(data))
{
[... snip ...]
}
You load the whole File into a memory stream, this requires 576MB.
[... snip ...]
using (Bitmap src = new Bitmap(memoryStream)) // <-- exception
{
[... snip ...]
}
[... snip ...]
You load the whole stream contents into a bitmap, this requires at least another 576MB (depending on how much memory the bitmap requires per pixel, should be at least 4, propably more). At that point you have the image twice in memory which seriously hurts for such big images.
You can reduce the memory footprint by getting rid of the memory stream and loading the bitmap directly from the file stream.
Another solution would be to load only a part of the bitmap and load the other parts on-demand (much like google maps), but i can't help you with that solution, might require reading the bitmap manually.

Not a complete answer to your question, but you are probably better of using a library like ImageMagick.NET

MemoryStream is implemented using an array of bytes that stores the data. If you read more data than the array can hold, a new array of double size is allocated and the bytes copied from one array to the other.
Since you apparently know how much data you're going to need, you can allocated the correct size up front and thus avoid the resizing.
However, once you reach a certain size you will run out of memory. .NET imposes a 2 GB limit on a single object (even on 64 bit), so the internal array in MemoryStream will never be able to grow beyond that. If your image is larger than that you'll get an out of memory exception.