Error from image_deskew(): Magick: TIFF: negative image positions unsupported about magick HOT 12 CLOSED

Could you add some example code? What does repage do?

from magick.

It removes/resets the virtual canvas meta-data from the images.

I've read that it solves the solution here: https://www.imagemagick.org/discourse-server/viewtopic.php?t=25961

from magick.

Do you have an example image/code that I can use to test this?

from magick.

I also added an magick::image_ocr() wrapper.

from magick.

Sure, this is the image: https://github.com/ShirinG/blog_posts_prep/blob/master/ocr/beispiel_scan.png

And the code:

library(tidyverse)
library(tesseract)
library(magick)

image <- image_read("https://raw.githubusercontent.com/ShirinG/blog_posts_prep/master/ocr/beispiel_scan.png")

image_bearb <- image %>%
  image_scale("x2000") %>%                        # rescale
  image_background("white", flatten = TRUE) %>%   # set background to white
  image_trim() %>%                                # Trim edges that are the background color from the image.
  image_noise() %>%                               # Reduce noise in image using a noise peak elimination filter.
  image_enhance() %>%                             # Enhance image (minimize noise)
  image_normalize() %>%                           # Normalize image (increase contrast by normalizing the pixel values to span the full range of color values).
  image_contrast(sharpen = 1) %>%                 # increase contrast
  image_deskew(treshold = 40)                     # deskew image -> creates negative offset in some scans == error

text_2 <- ocr(image_bearb,
              engine = tesseract(language = "deu"))

from magick.

Hmm it could actually be a problem with the dictionary. Do you get the same error if you don't set the language parameter?

from magick.

Yes, I do get the same error when I use

text_2 <- ocr(image_bearb)

from magick.

Hmm I can't reproduce the problem here. But it is probably caused by the fact that I use tiff to get the image from magick to tesseract format but maybe I should change that.

Could you try if the problem disappears when you save as png instead?

image_write(image_bearb, "test.png", format = "png")
tesseract::ocr("test.png", engine = tesseract(language = "deu"))

from magick.

Yeah, this works. So, there is no way to have one consecutive pipeline for the whole analysis?

from magick.

It will be fixed in the next version.

from magick.

Thank you!

from magick.

This should be fixed in the new version on CRAN.

Note that the magick package now also has a convenience wrapper for tesseract so you can pipe to image_ocr(language = "deu") without specifically loading the tesseract package.

from magick.