Adjusting image quality when resizing

Question

User768014394 posted

I'm trying to refactor some image resizing code to make it more efficient.  I have the resizing working fine, however the resulting image's bytes are a bit larger than the original code's results.

I have the following code

private Bitmap ResizeImage(Bitmap originalImage, int width, int height)
{
	Bitmap finalImage = new Bitmap(width, height);

	using (Graphics graphics = Graphics.FromImage(finalImage))
	{
		Rectangle rectangle = new Rectangle(0, 0, width, height);

		graphics.CompositingQuality = CompositingQuality.Default;
		graphics.SmoothingMode = SmoothingMode.None;
		graphics.InterpolationMode = InterpolationMode.Bilinear;

		graphics.DrawImage(originalImage, rectangle);
	}

	const int quality = 40;

	ImageCodecInfo codec = RetrieveCodec("image/jpeg");

	using (EncoderParameters codeParams = new EncoderParameters())
	{
		using (EncoderParameter p = new EncoderParameter(Encoder.Quality, quality))
		{
			using (MemoryStream bmpStream = new MemoryStream())
			{
				codeParams.Param[0] = p;

				// Exception occurs here
				finalImage.Save(bmpStream, codec, codeParams);
			}
		}
	}

	return finalImage;
}

private ImageCodecInfo RetrieveCodec(string mimeType)
{
	ImageCodecInfo[] codecs = ImageCodecInfo.GetImageEncoders();

	foreach (ImageCodecInfo codec in codecs)
	{
		if (codec.MimeType == mimeType)
			return codec;
	}

	return null;
}

 

When I call finalImage.Save(bmpStream, codec, codeParams), I get an Argument exception.  What am I doing wrong?  I checked the arguments and it seems everything has the right values.

Answer 1

User768014394 posted

Figured it out.  Apparently I wasn't copying the bitmap with "originalImage = resizedBmp;"

private Bitmap ResizeImage(Bitmap originalImage, int width, int height)
{
	using (Bitmap resizedBmp = new Bitmap(width, height))
	{
		using (Graphics g = Graphics.FromImage(resizedBmp))
		{
			g.CompositingQuality = CompositingQuality.Default;
			g.SmoothingMode = SmoothingMode.None;
			g.InterpolationMode = InterpolationMode.Bilinear;

			Rectangle bounds = new Rectangle(0, 0, width, height);
			g.DrawImage(originalImage, bounds);
		}

		originalImage = resizedBmp;

		const int quality = 40;

		ImageCodecInfo codec = RetrieveCodec("image/jpeg");

		using (MemoryStream bmpStream = new MemoryStream())
		{
			using (EncoderParameters codecParams = new EncoderParameters())
			{
				using (EncoderParameter p = new EncoderParameter(Encoder.Quality, quality))
				{
					codecParams.Param[0] = p;
					resizedBmp.Save(bmpStream, codec, codecParams);
				}
			}

			return new Bitmap(bmpStream);
		}
	}
}

Answer 2

User768014394 posted

That wasn't it.  I'm still getting the same exception.  The code changes I made were not being executed...

Answer 3

User768014394 posted

Here's what that problem was.  I am not sure why it is happening though.

I was declaring const int quality = 40.  When I created my EncoderParameter "p", the codecParams[0].Type was set to EncoderParameterValueType.ValyeTypeByte and that was apparently causing the argument exception.  One of two changes fixes it:

1. Instead of declaring quality as an integer constant, declare it as an integer variable
2. Or declare quality as a long constant

If I do either, the codecParams[0].Type is set to EncoderParameterValue.Type.ValueTypeLong.

Why would declaring it as an integer constant try to make it a byte and declaring it as an integer variable not?