This question already exists:
Application about C# image quality [closed]
Closed 2 years ago.
I am developing an application with C#. In the application I developed, I take the picture file with the user selection and transfer the picture onto the form.
After this step, I want to make changes to the bitmap of the picture. For example resizing the image in small sizes and converting to 1bpp color format.
I can do these now, I can resize the image and convert it to 1bpp color format, but at this point I think I have quality problems.
For example, when I take a screenshot of a text and send it to the program, I see that the letters in the text are unclearly bad when I view it in the resized 1bpp color format.
I show the algorithms I used and the screenshot of the application:
Bitmap bmp = new Bitmap(ResizeImage.filePath);
Bitmap bmpOriginalRGB = Helper.ImageResize(bmp, 512, 384);
pcbox1.Image = bmpOriginalRGB;
Bitmap bmpResizeRGB = Helper.ImageResize(bmp, ResizeWidth, ResizeHeight);
pcbox2.Image = bmpResizeRGB;
Bitmap bmpResize1BPP_1 = Helper.ConvertTo1BppImage(Helper.ImageResize(bmp, ResizeWidth, ResizeHeight));
pcbox3.Image = bmpResize1BPP_1;
Bitmap bmpResize1BPP_2 = bmpResizeRGB.Clone(new Rectangle(0, 0, ResizeWidth, ResizeHeight), PixelFormat.Format1bppIndexed);
pcbox4.Image = bmpResize1BPP_2;
Bitmap oledBitmap = bmpResize1BPP_2;
and
public static Bitmap ImageResize(Image image, int width, int height)
{
var destRect = new Rectangle(0, 0, width, height);
var destImage = new Bitmap(width, height);
destImage.SetResolution(image.HorizontalResolution, image.VerticalResolution);
using (var graphics = Graphics.FromImage(destImage))
{
graphics.CompositingMode = CompositingMode.SourceCopy;
graphics.CompositingQuality = CompositingQuality.HighQuality;
graphics.InterpolationMode = InterpolationMode.HighQualityBicubic;
graphics.SmoothingMode = SmoothingMode.HighQuality;
graphics.PixelOffsetMode = PixelOffsetMode.HighQuality;
using (var wrapMode = new ImageAttributes())
{
wrapMode.SetWrapMode(WrapMode.TileFlipXY);
graphics.DrawImage(image, destRect, 0, 0, image.Width, image.Height, GraphicsUnit.Pixel, wrapMode);
}
}
return destImage;
}
Here's a first screenshoot
Here's a second screenshoot
Here's a third screenshoot
For example, as in the second screen image, the shadow of the woman appears as black. And I don't want that. I want to see a clearer black and white image.
Do you think it is possible to further improve this picture quality?
Edit: Sorry for translate..
I don't believe your problem is the quality on resizing, it's rather the conversion from greyscale to black and white.
When resizing, the color of the adjacent pixels are averaged together to get a new color approximating the set of pixels. So going from a 200x200 pixel image to a 100x100 pixel image, a set of 4 pixels becomes a single pixel. It will still look fairly good, but the solid black text will become a series of gray pixels. The exact formula will vary by the interpolation method used. But when resizing a nice piece of black and white text, the text will end up lighter or more grey (which will be important later)
However, when going from greyscale (or full color) to black and white each pixel ends up being either black and white, there are no other options, but there are different algorithms used to decide which pixels end up black or white, often called dithering. (https://en.wikipedia.org/wiki/Dither)
Your first b/w image appears to use some form of error diffusion, quite possibly Floyd-Steinberg. It tends to work fairly well on real wold images and turns grey areas into spaced out black pixels visually approximating the greyness of the area they fall in.
Your second b/w image appears to be a simple threshold algorithm. Basically pixels darker than a certain color end up black, and all the rest end up white. You can adjust the image by simply setting which color is used as the threshold. Often this works well with text, but you will need to adjust the value used as the threshold, but I don't know if the libraries you are using allow for this or not. I have found what works well for programmatically selecting the threshold is to total up how many pixels there are of each color, and then assume some percent will be black (text tends to be mostly white space), then choose the threshold that gives you that number of black pixels.
And there are many other dithering algorithms that you can try, as well as edge detection algorithms. You can also try adjusting the contrast of the image before converting to b/w.
However, at the end of the day, when resolution is low (pixels per character), it may not be possible to easily convert them to b/w and have them still be readable (just try to fax small text on a fax machine in standard mode). Remember your resize removed a lot of information and the conversion from 8 bit to 1 bit removed another 87% of the information in the resized picture.
Related
Hello Stackoverflow Community.
I currently work on an application which has a cursor region magnifyer feature, for the user to pick a color.
However, i've the problem that the Picturebox has white edges on the right and bottom, even though the image captured is only one dark color.
The captured screen is 10x10 pixel and the Picturebox is 80x80 pixel and it's SizeMode set to StretchImage.
I checked if the raw captured image maybe contains this edges already by saving it and checking it in Photoshop. But the raw captured image is fine, so it must be something weird with the Picturebox.
Here you can see how it looks (The mousecursor and the capture-region[green rectangle] are just painted on the screenshot for demonstration, as i was not able to screen my cursor, and yes - the green region is painted way too large, it should be just 10x10 pixel ^^)
Has it maybe something to do with the way the Image get's stretched internally? If so, is there any, not too complicated way around it?
I appreciate any help in advance :)
//Edit: I think it's actually about the stretching
I found this topic Image after resize has white border but i don't know what the GetSize() Method is, or where it is (from)
Have a great day!
I found a solution for the edges here, as you can see in the first screen. However, in the second screen you can see a 2x2 pixel capture, stretched to 80x80 pixel with the found "solution", will mix colors as it's actually 80x80 pixel in the end, while i want to have it displaying the raw pixels, so, 3 black ones and one white one in this case. Now i am even more stuck ^^
I guess i should instead read the pixel colors of the single pixels from captured image and set the colors in the preview picturebox, or make 4 panels in the case of just 2x2 pixels
public new Image Resize(Image image, int targetWidth, int targetHeight)
{
var resizedImage = new Bitmap(targetWidth, targetHeight);
using (var graphics = Graphics.FromImage(resizedImage))
{
graphics.InterpolationMode = InterpolationMode.HighQualityBicubic;
var attributes = new ImageAttributes();
attributes.SetWrapMode(WrapMode.TileFlipXY);
var destination = new Rectangle(0, 0, targetWidth, targetHeight);
graphics.DrawImage(image, destination, 0, 0, image.Width, image.Height, GraphicsUnit.Pixel, attributes);
}
return resizedImage;
}
//Edit:
Okay i think i'm gonna use Get/SetPixel or Panels(for small captures), as i did here on the right, the left is still the 80x80 Picturebox
I'm making a tool that does some graphical editing to some bitmaps. I have some 8bppIndexed bitmaps with an existing color palette that I have to convert to 24bppRgb to allow me to draw on it and edit colors of pixels based on some options the user selected. Bitmap graphic is the original 8bpp .bmp file:
//Get Bitmap from file
Bitmap graphic = new Bitmap(source);
//Get palette and pixel format
ColorPalette pal = graphic.Palette;
PixelFormat pix = graphic.PixelFormat; //Format8bppIndexed
//Create datagraphic in 24bppRgb to allow for editing
PixelFormat datapix = PixelFormat.Format24bppRgb;
Bitmap datagraphic = graphic.Clone(new Rectangle(0, 0, graphic.Width, graphic.Height), datapix);
After drawing and editing Bitmap datagraphic, I'm trying to convert it back to 8bppIndexed with the same color palette as Bitmap graphic. I can do that like this:
//Convert back to graphic pixel format
datagraphic = datagraphic.Clone(new Rectangle(0, 0, datagraphic.Width, datagraphic.Height), pix); //Format8bppIndexed
//Apply color palette
datagraphic.Palette = pal;
However, after cloning datagraphic with the same pixel format as graphic, it creates an entirely new color palette. The colors in datagraphic are then all incorrect after applying ColorPalette pal. They only match by their index number between the two palettes. Is there another way of doing this that preserves the colors?
And yes, the bitmaps need to be 8bppIndexed with the custom palette. I'm trying to avoid the need to go through Photoshop to change the color index of all the end result data graphics with the correct palette.
I should say that I'm still fairly new to C#. I hope I was clear as to what I'm trying to do. I appreciate any help here. Thanks in advance.
Converting an 8-bit image to 24bpp is trivial: you just paint it on a new image of the same dimensions and the preferred pixel format.
public static Bitmap PaintOn24bpp(Image image)
{
Bitmap bp = new Bitmap(image.Width, image.Height, PixelFormat.Format24bppRgb);
using (Graphics gr = Graphics.FromImage(bp))
gr.DrawImage(image, new Rectangle(0, 0, bp.Width, bp.Height));
return bp;
}
This works with any source image no matter its pixel format, though for images containing transparency you may want to preprocess them to fill any transparent areas, because I don't know if painting with transparency taken into account will work if the image you're drawing on doesn't support alpha.
The reverse, to convert a high-colour image to indexed, is a lot trickier, but if you have the palette, it is certainly possible. The main problem in this case is that using Graphics doesn't work, since indexed graphics don't have coloured pixels; they have a coloured palette, and pixels that just reference those colours.
Editing 8-bit data generally boils down to building a byte array containing the image data, and converting that to an image with the palette slapped onto it. The actual conversion is done by making a new 8bpp Bitmap, opening its backing byte array with the LockBits function, and copying your new data into it with Marshal.Copy.
After that, the final missing link is a method to match your image's coloured pixels to their closest palette match, so you can then take the array of resulting matches and bake it into an 8bpp image. This is generally done by calculating the Pythagorean distance (in 3D RGB colour space) of your colour to each of the palette colours, and then taking the index of the colour for which that distance is the smallest.
The whole process, from start to end, is explained in detail in this answer:
A: How to convert a colored image to a image that has only two predefined colors?
This deals with 2 colours, but the method for dealing with 2 or with 256 colours is completely identical, since the end result in both cases was an 8bpp image.
It could be optimised for dealing with 24bpp images directly instead of converting the input to 32bpp, but that's probably not worth the effort, especially since the method posted there works with absolutely any input.
I am trying to precisely and predictably scale an image in C# to a different resolution, both up and down. When I open the resulting images with external tools such as Gimp, the results are not satisfying with my current settings.
public Image Square(Image image, int res) {
Bitmap sq = new Bitmap(res, res, image.PixelFormat);
Graphics canvas = Graphics.FromImage(sq);
canvas.CompositingQuality = CompositingQuality.HighQuality;
canvas.SmoothingMode = SmoothingMode.None;
canvas.InterpolationMode = InterpolationMode.Bicubic;
canvas.DrawImage(sq, 0, 0, res, res);
return sq;
}
The results are okay when scaling down (but far from perfect), but there are side-effects when scaling up:
This picture has a resolution of 2x2 pixels. The alpha channel is set to opaque for all pixels.
After scaling it to 4x4 pixels, this is the result:
Apparently, the C# graphics library introduced transparency while scaling the picture. This method should still work if the given image has transparent pictures, so removing the alpha channel is not an option.
Similarly, when scaling pictures down, there are problems at the edges of the resulting images as well, usually either very dark or transparent.
Is there any way to circumvent this behavior?
Edit: I already tried NearestNeighbor for downscaling only, but it results in this:
Edit 2: With WrapMode.TileFlipXY, the transparent edge is gone, but red only makes up 25% of the image instead of 50% as it should be:
You're asking for bicubic interpolation and you're getting it. What you want is the "nearest neighbor" option as outlined in the docs:
canvas.InterpolationMode = InterpolationMode.NearestNeighbor;
A way to avoid edge artifacts is to wrap the image:
using (ImageAttributes wrapMode = new ImageAttributes())
{
wrapMode.SetWrapMode(WrapMode.TileFlipXY);
g.DrawImage(input, rect, 0, 0, input.Width, input.Height, GraphicsUnit.Pixel, wrapMode);
}
Direct copy/paste from:
Ghost-borders ('ringing') when resizing in GDI+
I believe you need to combine NearestNeighbor interpolation with Half pixel offset. As pointed out in a similar question here.
canvas.InterpolationMode = InterpolationMode.NearestNeighbor;
canvas.PixelOffsetMode = PixelOffsetMode.Half;
I have two overlapping pictureboxes.The images of both picture boxes have some transparent pixels.I want to see the bottom picture box through the transparent pixels of the overlapping picture box.
I tried setting the background color of both picture boxes as transparent.But it just sets the back color of the picture box to the background color of the form.
Clearly you are using Winforms. Yes, transparency is simulated by drawing the pixels of the Parent. Which is the form, you only see the form pixels, stacking effects don't work. There's a KB article that shows a workaround for this. It is painful. Another approach is to not use PictureBox controls but just draw the images in the form's Paint event.
Consider WPF, it has a very different rendering model that easily supports transparency.
Solutions to that problem might be various, and it mainly depends on your skills and amount of work will depend on kind of images you're dealing with. For example if images are always same resolution, size and overlapping image supports transparency you could try to do manipulation of two Image objects and draw one over another, then display it in PictureBox. Or if you will need to do it multiple times in various places of your app you could even consider creating your own UserContriol.
Code in answer of this question, method ResizeImagein particular, show how to create resized, good quality image, all you need it is to change it a little. Make it to get two Images as input parameters, and change it to draw one image over another.
Changes might look like this
public static Bitmap CombineAndResizeTwoImages(Image image1, Image image2, int width, int height)
{
//a holder for the result
Bitmap result = new Bitmap(width, height);
//use a graphics object to draw the resized image into the bitmap
using (Graphics graphics = Graphics.FromImage(result))
{
//set the resize quality modes to high quality
graphics.CompositingQuality = System.Drawing.Drawing2D.CompositingQuality.HighQuality;
graphics.InterpolationMode = System.Drawing.Drawing2D.InterpolationMode.HighQualityBicubic;
graphics.SmoothingMode = System.Drawing.Drawing2D.SmoothingMode.HighQuality;
//draw the images into the target bitmap
graphics.DrawImage(image1, 0, 0, result.Width, result.Height);
graphics.DrawImage(image2, 0, 0, result.Width, result.Height);
}
//return the resulting bitmap
return result;
}
And use it, for example, like this:
pictureBox1.Image = CombineAndResizeTwoImages(Image.FromFile("c:\\a.png"), Image.FromFile("c:\\b.png"), 100,100);
But that its only example, and you must tune it up to your needs.
Good luck.
If it's one PictureBox inside another, you can use:
innerPictureBox.SendToBack();
innerPictureBox.Parent = outerPictureBox;
I'm trying to scale down a Bitmap using GDI+ by doing the following:
Bitmap newImage = new Bitmap(NewWidth, NewHeight, Im.PixelFormat);
Graphics g = Graphics.FromImage(newImage);
g.CompositingQuality = System.Drawing.Drawing2D.CompositingQuality.HighQuality;
g.InterpolationMode = System.Drawing.Drawing2D.InterpolationMode.HighQualityBicubic;
g.SmoothingMode = System.Drawing.Drawing2D.SmoothingMode.HighQuality;
g.ScaleTransform(0.1, 0.1); // 10%
g.DrawImage(Im, 0, 0, Im.Width, Im.Height);
Im is the original image, NewWidth and NewHeight are 10% or the original image. I've tested this on a 1000x1000 image (shrinking it down to 100x100)
The scaling is done correctly with high quality as promised but for some reason there is a gray border on the left, right and top borders (none on the bottom).
I assume this is due to the fact the all the image borders are white and the color "outside" of the bitmap is by default black so some of the default black get mixed into the scaling interpolation.
I looked for a way to set the default background color to white (white will do just fine) but couldn't find it anywhere..
My alternative is to pad the border with a white frame, scale the image down and the crop it but I was wondering if there is a simpler and less CPU consuming way?
Any ideas?
Well. After some more digging I found it..
System.Drawing.Imaging.ImageAttributes Att = new System.Drawing.Imaging.ImageAttributes();
Att.SetWrapMode(System.Drawing.Drawing2D.WrapMode.Clamp, System.Drawing.Color.White);
g.DrawImage(Im, new Rectangle(0,0,Im.Width,Im.Height), 0, 0, Im.Width, Im.Height, GraphicsUnit.Pixel, Att);
Try adding an alpha overlay with an extra transparent pixel or two on all sides. This should get a better result than using either an explicit or implied solid color frame. Though converting to RGBA and adding the frame has a higher execution cost, if you are really interested in high quality and don't want to switch graphics libraries it may be the way to go.