Improved clarity of page images

[deahhh]

I'm not sure if this is a problem.
Does the library only extract the data contained in the pdf file?
Does the pdf have a protocol for blurring the data and displaying it?
Evince can display very clear content (not reproducible, probably images, using fitz can get the image of the base64 string), but use fitz to extract the page image file is very blurred, is Evince made digital filtering process?

I'm very sorry, I asked Evince, but I didn't get a reply, so I didn't have a solution, so I thought I'd try my luck here.

Does this library have plans for super-resolution processing of pdf displays?

[JorjMcKie]

This is no issue, but a question, so I converted your post.

Several comments in your post are hard to understand:

• what do you mean with "blurred", "blurring" data?
• how comes base64 encoding into play here?
• do you mean making an image from a page or extracting images shown on a page?

I am confused.

[deahhh]

Does it come from pdf protocols which limit it to show? Or come from Evince who may be doing extra work to make it readable?

nice, it must be some tech used by Evince, which make the showing in it more clear and readable.

[JorjMcKie]

what do you mean 'not created by me'? Please tell me the instruction that delivered the image

[JorjMcKie]

it's not created by me, but read from the file with pymupdf.
it's image file, in string.

Again my question: what did you do to read from the file?

• doc.extract_image()?
• page.get_pixmap()?
• or what else?

[deahhh]

Thanks again!
page.get_pixmap()
I should give this information at beginning. So sorry.

The pdf

code:

import fitz
import re
import os

file_path = r'1.pdf'       # PDF 
dir_path = r'./images'  

def pdf2image1(pdf_path, pic_path):
    checkIM = r"/Subtype(?= */Image)"
    doc = fitz.open(pdf_path)

    for i in range(1, doc.page_count):
        pg = doc.load_page(i).get_svg_image()
        
        pix = doc[i].get_pixmap()
        pix.save(os.path.join(pic_path, f"page_{i}.png"))

pdf2image1(file_path, dir_path)

[deahhh]

@JorjMcKie
Thank you very much!
I've got the origin image now. Inspired by your example code

[JorjMcKie]

I should give this information at beginning. So sorry.

Indeed you should!

If you look at the documentation, you will find several options to increase the resolution of a pixmap. The easiest to use is parameter dpi: page.get_pixmap(dpi=300) for example creates an image with 300 dots per square inch.

[JorjMcKie]

And another comment:
When using page.get_pixmap(), you are creating an image. It is now clear that your previous statement "it's not created by me, but read from the file with pymupdf" was simply wrong.

[deahhh]

Yes.
It's the name get_xxx made me consider get_pixmap as nothing to do with create.
Stereotypes of thinking made me consider that there is a show do something like draw_pixmap.

Thanks again!
I think it's great to use pymupdf do something interesting.