Auto crop PDF based on contents

Aspose.PDF Product Family
1

Hello


I am trying to crop the PDF document based on contents in a file. I can do it with an image but cannot achieve the same with a PDF. I am trying to stay with PDF as the quality seems clearer
than PNG/GIF, etc

Regards,
Simon
2

Hi Simon,


Thanks for contacting support.

To remove or trim the white space around a PDF page, set the CropBox for that particular page. To determine the correct value of the CropBox coordinates, first you need to identify the position of objects on the page. You can use following code snippet as a sample to trim white spaces around the page.

C#

// Load an existing PDF files
Document doc = new Document(dataDir + “input.pdf”);
// Render the page to image with 72 DPI
PngDevice device = new PngDevice(new Resolution(72));
using (MemoryStream imageStr = new MemoryStream())
{
device.Process(doc.Pages[1], imageStr);
var bmp = (System.Drawing.Bitmap)System.Drawing.Bitmap.FromStream(imageStr);
System.Drawing.Imaging.BitmapData imageBitmapData = null;
// Determine white areas
try
{

imageBitmapData = bmp.LockBits(new System.Drawing.Rectangle(0, 0, bmp.Width, bmp.Height),
System.Drawing.Imaging.ImageLockMode.ReadOnly, System.Drawing.Imaging.PixelFormat.Format32bppRgb);
Aspose.Pdf.Rectangle prevCropBox = doc.Pages[1].CropBox;
int toHeight = bmp.Height;
int toWidth = bmp.Width;
int? leftNonWhite = null;
int? rightNonWhite = null;
int? topNonWhite = null;
int? bottomNonWhite = null;
for (int y = 0; y < toHeight; y++)
{
byte[] imageRowBytes = new byte[imageBitmapData.Stride];
// Copy the row data to byte array
if (IntPtr.Size == 4)
System.Runtime.InteropServices.Marshal.Copy(new IntPtr(imageBitmapData.Scan0.ToInt32() + y * imageBitmapData.Stride), imageRowBytes, 0, imageBitmapData.Stride);
else
System.Runtime.InteropServices.Marshal.Copy(new IntPtr(imageBitmapData.Scan0.ToInt64() + y * imageBitmapData.Stride), imageRowBytes, 0, imageBitmapData.Stride);

int? leftNonWhite_row = null;
int? rightNonWhite_row = null;
for (int x = 0; x < toWidth; x++)
{
if (imageRowBytes[x * 4] != 255
|| imageRowBytes[x * 4 + 1] != 255
|| imageRowBytes[x * 4 + 2] != 255)
{
if (leftNonWhite_row == null)
leftNonWhite_row = x;
rightNonWhite_row = x;

}

}

if (leftNonWhite_row != null || rightNonWhite_row != null)
{
if (topNonWhite == null)
topNonWhite = y;
bottomNonWhite = y;
}

if (leftNonWhite_row != null
&& (leftNonWhite == null || leftNonWhite > leftNonWhite_row))
{
leftNonWhite = leftNonWhite_row;
}
if (rightNonWhite_row != null
&& (rightNonWhite == null || rightNonWhite < rightNonWhite_row))
{
rightNonWhite = rightNonWhite_row;
}
}

leftNonWhite = leftNonWhite ?? 0;
rightNonWhite = rightNonWhite ?? toWidth;
topNonWhite = topNonWhite ?? 0;
bottomNonWhite = bottomNonWhite ?? toHeight;
// Set crop box with correction to previous crop box
doc.Pages[1].CropBox =
new Aspose.Pdf.Rectangle(
leftNonWhite.Value + prevCropBox.LLX,
(toHeight + prevCropBox.LLY) - bottomNonWhite.Value,
rightNonWhite.Value + doc.Pages[1].CropBox.LLX,
(toHeight + prevCropBox.LLY) - topNonWhite.Value
);
}
finally
{
if (imageBitmapData != null)
bmp.UnlockBits(imageBitmapData);
}
}
dataDir = dataDir + “TrimWhiteSpace_out.pdf”;
// Save the updated document

doc.Save(dataDir);

If you still face any issue or need further assistance, please feel free to contact us.

Best Regards,