extract image problem

[kar9999]

Description of the bug

When using PyMuPDF's
get_ text("dict") method to
extract electronic sealvector
graphics, the backgrouind is
black. When the vector
graphics themselves are also
black, they become
indistinguishable. Is there a way to make the background
transparent?

How to reproduce the bug

1

PyMuPDF version

1.26.7

Operating system

Linux

Python version

3.10

[JorjMcKie]

Without an example file we cannot deal with this.

[kar9999]

example.pdf
This is a sample file. The electronic seal is like an image of a persimmon, and its background is transparent. However, when extracting it using the get_text("dict") method, the extracted image has a black background. Is there any way to configure or modify the get_text("dict") function so that the extracted image has a white background instead?

The inserted image of a persimmon with a transparent background.：

The image extracted via get_text("dict")：

[JorjMcKie]

Extract both, base and mask binary from the dictionary like so:

import pymupdf

print(f"{pymupdf.version=}")
doc = pymupdf.open("example.pdf")
page = doc[0]
# extract all blocks ensuring that we also see partially shown images
blocks = page.get_text("dict", clip=pymupdf.INFINITE_RECT())["blocks"]
images = [b for b in blocks if b["type"] == 1]  # select any images
print(f"Found {len(images)} images on the first page.")
img = images[0]
base = img["image"]  # the base image binary
pixbase = pymupdf.Pixmap(base)
mask = img.get("mask")  # the mask binary, if any
if mask:
    print("Image 0 has a mask.")
    pixmask = pymupdf.Pixmap(mask)
else:
    pixmask = None
# combine base and mask if any
pixfinal = pymupdf.Pixmap(pixbase, pixmask) if pixmask else pixbase
pixfinal.save("extracted_image.png")