Deep/Fast Removal not working correctly

[ZingZing001]

Thanks to everyone who has been using my program lately, and hopefully, this small tool has helped you out.

Just some Q&A's about the program based on recent activities:

1. The fast removal has taken away not just the watermark but also some other elements from the pdf!

This is probably due to my algorithm which removes all ocg layers within the pdf. To solve this issue, use any pdf editor and check out the layering page, identify the name for the watermark layer, and modify it within the base code as suggested by chatgpt:

def remove_specific_ocg(pdf_path, output_path, ocg_layer_name):
    with pikepdf.Pdf.open(pdf_path, allow_overwriting_input=True) as pdf:
        # Check if /OCProperties exists in the document catalog
        if "/OCProperties" in pdf.Root:
            ocgs = pdf.Root["/OCProperties"].get("/OCGs", None)
            if ocgs:
                # Iterate over all OCG objects to find the one with the given name
                ocg_to_remove = None
                for ocg in ocgs:
                    if ocg.get("/Name") == ocg_layer_name:
                        ocg_to_remove = ocg
                        break
                
                if ocg_to_remove:
                    print(f"Found OCG layer '{ocg_layer_name}'. Removing...")
                    # Remove the OCG from the /OCGs array
                    ocgs.remove(ocg_to_remove)

        # Iterate over all pages to clean references to the specific OCG
        for page_number, page in enumerate(pdf.pages, start=1):
            print(f"Processing page {page_number}...")
            
            # Update /Contents to remove references to the specific OCG
            if "/Contents" in page:
                content_obj = page["/Contents"]
                if isinstance(content_obj, pikepdf.Array):
                    new_contents = pikepdf.Array()
                    for obj in content_obj:
                        stream = obj.read_bytes()
                        # Remove commands related to the specific OCG
                        if ocg_layer_name.encode() in stream:
                            stream = stream.replace(ocg_layer_name.encode(), b"")
                        new_obj = pdf.make_stream(stream)
                        new_contents.append(new_obj)
                    page["/Contents"] = new_contents
                elif isinstance(content_obj, pikepdf.Stream):
                    stream = content_obj.read_bytes()
                    if ocg_layer_name.encode() in stream:
                        stream = stream.replace(ocg_layer_name.encode(), b"")
                    page["/Contents"] = pdf.make_stream(stream)

            # Remove XObjects tied to the specific OCG
            if "/Resources" in page:
                resources = page["/Resources"]
                if "/XObject" in resources:
                    xobjects = resources["/XObject"]
                    keys_to_remove = []
                    for key, obj in xobjects.items():
                        if obj.get("/OC") == ocg_to_remove:
                            keys_to_remove.append(key)
                    for key in keys_to_remove:
                        print(f"Removing XObject '{key}' tied to OCG layer '{ocg_layer_name}'.")
                        del xobjects[key]

            # Remove annotations tied to the specific OCG
            if "/Annots" in page:
                annotations = page["/Annots"]
                new_annotations = pikepdf.Array([annot for annot in annotations if annot.get("/OC") != ocg_to_remove])
                page["/Annots"] = new_annotations

        # Save the cleaned PDF
        pdf.save(output_path)
        print(f"Saved cleaned PDF without the specific OCG '{ocg_layer_name}' to {output_path}")
        return f"Successfully removed the OCG '{ocg_layer_name}' and saved to {output_path}"

To test this, use the variables below:

pdf_path = "input.pdf"
output_path = "output.pdf"
ocg_layer_name = "Watermark Layer"  # Replace with the actual name of the OCG layer
result = remove_specific_ocg(pdf_path, output_path, ocg_layer_name)
print(result)

This should remove only the ocg layer you've specified.

2. Deep removal won't work properly for coloured pdf?

If the watermark is "Underneath" the background colour, simply use a colour dropper to gather the RGB value for the watermark, and adjust the value for the RGB range and make sure to avoid that colouring range.
Such as:
RGB for watermark: RGB(245, 200, 210)
Use the values below can exclude the watermark, also make sure to change the HSV values and make sure you don't alter the saturation/value value, you can try set it to 0, <200 accordingly.

E.g. (Check the comment section within the base code)

For Blue JB WaterMark
      (img_array[:, :, 0] < 200) &  # Red channel threshold increased
      (img_array[:, :, 1] < 200) &  # Green channel threshold increased
      (img_array[:, :, 2] < 175)    # Blue channel threshold increased
HSV value:
      (0,250)

3. The program is not working from the Release Version:

The program within release is merely just a sneak-peek of what my program looks like, and it will only remove a specific type of watermark tested by me. You are strongly encouraged & advised to compile it using the python compiler yourself, as this would allow you to alter with the value (rgb,hsv) to make the program that fits YOUR needs.
PLEASE READ THE CONFIGURATION WITHIN THE README FILE BEFORE POSTING ANYTHING IN issues