We Keep Coding

Tile an image in PDF using iText7

I would like to be able to fill graphics objects using the tiling pattern functionality that pdf offers. For example, I would like to be able to draw something like this:
iText7 has a few objects related to patterns that could be useful, but I am having trouble figuring out how to use them and it is exceedingly difficult to find examples of similar code online.
iText7 provides the following classes that may be useful:
PdfPattern.Tiling
PatternColor
PdfPatternCanvas
It looks like you should be able to create a PdfPattern.Tiling object which references an image in some way and then create a PatternColor from that tiling object. Then you can set your canvas' fill color to the PatternColor you just created. An example of a function that does this is:
private void SetImageTilingFill(PdfCanvas canvas, PdfImageXObject img)
{
PdfPattern.Tiling tiling = new PdfPattern.Tiling((float)Inches2Points(img.GetHeight() / 96), (float)Inches2Points(img.GetWidth() / 96)); // create tiling object with width and height the size of the img
tiling.GetResources().AddImage(img);// add the image as a resource?
canvas.SetFillColor(new PatternColor(tiling)); // set fill color to PatternColor?
}
So far this approach has not been successful, my rectangle ends up solid black. Any suggestions would be much appreciated.

Using an image as a tile requires setting up the tile (it has a canvas to draw on). And then using it as a fill when drawing.
try (PdfWriter writer = new PdfWriter("tiling.pdf");
PdfDocument document = new PdfDocument(writer)) {
// creating image
ImageData imageData = ImageDataFactory.create("wavy.png");
PdfImageXObject imageXo = new PdfImageXObject(imageData);
// setting up the tile
PdfPattern.Tiling imageTile = new PdfPattern.Tiling(100, 100);
new PdfPatternCanvas(imageTile, document)
.addXObjectFittedIntoRectangle(imageXo, new Rectangle(0, 0, 100, 100))
.release();
PdfPage page = document.addNewPage();
PdfCanvas canvas = new PdfCanvas(page);
//using the tile
canvas.setFillColor(new PatternColor(imageTile));
canvas.rectangle(10, 10, 500, 900).fill();
canvas.release();
}

Xamarin.iOS Image on Image Watermark

How can I add a png image as a watermark to a larger image using Xamarin.iOS c# and save the output to the device?
I figured out the Xamarin.Android version from another question posted here.
Thanks in Advance!!

Using an image context, you can draw the original, then the watermark at the necessary location and obtain a new image from the context.
ImageContext example:
var originalImage = UIImage.FromBundle("buymore.jpg");
var watermarkImage = UIImage.FromFile("vs.png");
UIGraphics.BeginImageContextWithOptions(originalImage.Size, true, 1.0f);
originalImage.Draw(CGPoint.Empty);
watermarkImage.Draw(new CGRect(new CGPoint(200, 200), watermarkImage.Size));
var processedImage = UIGraphics.GetImageFromCurrentImageContext();
If your original and watermark images are the same size, you can use a CIFilter (CISourceOverCompositing) to "overlay" one image on top of another (assuming your watermark has a white or alpha background. This is my preferred method due to the speed.
CISourceOverCompositing example:
UIImage processedimage;
using (var filter = new CISourceOverCompositing())
{
filter.Image = new CIImage(UIImage.FromBundle("vs.png"));
filter.BackgroundImage = new CIImage(UIImage.FromBundle("buymore.jpg"));
processedimage = UIImage.FromImage(filter.OutputImage);
}

Tesseract OCR for .net not detecting text when white text is written on black rectangle and the border of black rectangle is white?

I am trying to extract a text from a png using tesseract. Tesseract doesn't extract text in case of following png.
The black rectangle in the image has white border(which is not visible because of website background).But, in case, I remove white border off the rectangle then Tesseract is able to detect the text.Is there any work around for this issue?
Also, if the decrease the border size then it does extract some text as:
I (31.04 I
Here is the code that I am using:
using (TesseractEngine ocr = new TesseractEngine(dataPath, "eng", EngineMode.TesseractOnly))
{
using (Pix p = Pix.LoadFromFile(filePath))
{
using (Pix img = p.Scale(2,3))
{
using (var page = ocr.Process(img))
{
string text = page.GetText();
Console.WriteLine(text);
}
}
}
}
By scaling I am able to extract following text:
I G1.04 I
However, if I increase the border size then even scaling the image has no effect.

using AForge.Imaging;
Grayscale grayFilter = new Grayscale(0.2125, 0.7154, 0.0721);
Bitmap grImage = grayFilter.Apply(image);
grImage.Save("./grey_image.png");
Pass this image path to your TesseractEngine.

Fillholes function in Aforge

I need to use Fillholes function of Aforge, it accepts binary image. I manipulated all pixels to black or white pixels using following code in c#:
bitmapimage.SetPixel(i, j, Color.FromArgb(255,255,255)); // for white pixel
bitmapimage.SetPixel(i, j, Color.FromArgb(0,0,0)); // for black pixel
But when I apply fillholes function to bitmap image, I get this exception:
"Source pixel format is not supported by the filter"
Kindly anyone help why I am getting this exception ... is bitmap image not converted to Binary by all using setpixel?

Just changing the pixel colors will not change the pixel format of your image.
You first need to make sure that you have a gray scale image using some gray scale filter, then make sure that the gray scale image is binary through some threshold filter. Once the image has been pre-processed using these steps, you may apply the FillHoles filter.
AForge.NET offers helper classes to merge several filters, so you can combine all three filters into one total filter using the FiltersSequence class.
Assuming that your original Bitmap image is named bitmap, you can then apply the fill holes filter for example like this:
var filter = new FiltersSequence(Grayscale.CommonAlgorithms.BT709,
new Threshold(100), new FillHoles());
var newBitmap = filter.Apply(bitmap);

AForge FillHoles Class
The filter allows to fill black holes in white object in a binary image. It is possible to specify maximum holes' size to fill using MaxHoleWidth and MaxHoleHeight properties.
The filter accepts binary image only, which are represented as 8 bpp images.
Sample usage:
C#
// create and configure the filter
FillHoles filter = new FillHoles( );
filter.MaxHoleHeight = 20;
filter.MaxHoleWidth = 20;
filter.CoupledSizeFiltering = false;
// apply the filter
Bitmap result = filter.Apply( image );
The above was found at http://www.aforgenet.com/framework/docs/html/68bd57bd-1fd6-6c4e-4500-ed4726bc836e.htm
You have to convert your bitmapImage to a binary image represented as an 8 bpp image. Here is one way to do it.
UnmanagedImage grayImage = null;
if (image.PixelFormat == PixelFormat.Format8bppIndexed)
{
grayImage = bitmapImage;
}
else
{
grayImage = UnmanagedImage.Create(image.Width, image.Height, PixelFormat.Format8bppIndexed);
}

Screen shoot quality worse than normal

I am trying to capture a part of my screen.
The problem is even if I use png to save the image the quality still worse than if I would just use normal print screen.
Here is the code I use:
//display a save file dialog for the user to set the file name
SaveFileDialog saveFileDialog = new SaveFileDialog();
saveFileDialog.Filter = "PNG (*.png)|*.png";
saveFileDialog.FilterIndex = 0;
saveFileDialog.RestoreDirectory = true;
//if the user proceed saving the picture
if (saveFileDialog.ShowDialog() == DialogResult.OK)
{
//simplify code with constant numbers for demo
//get the width of the panel we need the screenshoot off
int x = 10;
//get the height of the panel we need the screenshoot off
int y = 10;
//create the ractangle of the screenshoot panel
Rectangle rect = new Rectangle(x, y, 5, 5);
//create new bitmap
Bitmap bmp = new Bitmap(rect.Width, rect.Height, PixelFormat.Format32bppArgb);
Graphics g = Graphics.FromImage(bmp);
//get the screenshoot of the panel
g.CopyFromScreen(rect.Left, rect.Top, 0, 0, bmp.Size, CopyPixelOperation.SourceCopy);
string fileName = saveFileDialog.FileName;
if (!fileName.Contains(".png"))
fileName += ".png";
bmp.Save(fileName, ImageFormat.Png);
}
EDIT:
Example image form what I take with code:
Normal screenshot:
It does not look so different here, but it is worst.

The top image in your question was rescaled, smaller than the original. This is noticeable in images that contain fine detail, like the ClearType anti-aliasing pixels used to make the text more readable. When they get rescaled, the visual effect is ruined and text looks a lot worse.
It is entirely unclear why the image was rescaled, nothing in your code could cause that. Double-check by using the debugger to inspect the bmp.HorizontalResolution property, it should match the DPI of your video adapter. Simplest explanation that it was done by whatever image viewing program you used, perhaps to make the image fit the window. Try zooming out.

If it's possible to use a external library I suggest you FMUtils.Screenshot. Its available as a NuGet-package.
I just tried it and the quality is like the standard screenshot from windows. Here's a shortexample:
new ComposedScreenshot(new Rectangle(0, 0, 100, 100)).ComposedScreenshotImage.Save(#"PATH_TO_FILE\example-screenshot.png", ImageFormat.Png);
Hope this helps!

The pixel format you use only uses 8 bits each for the different color channels. You could try using PixelFormat64bppARGB to get 16 bits per color.
Resource on the PixelFormat enumeration: http://msdn.microsoft.com/en-us/library/system.drawing.imaging.pixelformat%28v=vs.110%29.aspx