问题
Experimenting with itext I am extracting both text and images from pdf files. For my purpose I build an html file using the text and images. The goal is to place the extracted images throughout the text rather than placing them at the end as I do currently.
After some research itext renderInfo.getImageCTM() appears to be just what I need, however the coordinates returned do not resemble the positions of some of the images when comparing to what's being displayed in adobe reader.
My MyImageRenderListener class has this
Matrix matrix = renderInfo.getImageCTM();
float x = matrix.get(Matrix.I31);
float y = matrix.get(Matrix.I32);
Which should give me the XY, top left hand corner position of each image, if I've understood it correctly.
Here is the pdf file I will be referring to
It's a 4 page PDF that contains either one or two images per page.
PAGE X Y
Page1 33.0 358.5
Page2 1st 321.7 419.9
Page2 2nd 41.2 182
Page3 1st 43.1 307.5
Page3 2nd 417 58.5
Page4 292.5 457.5
When comparing the coordinates output to the actual document the numbers do not seem to make sense. For example, On page3, the second image Y (58.5) is lower than the first image Y (307.5) on the same page. That would put image 2 on page 3 before image 1??
Also none of the Y and some of the X coordinates appear to be correct when comparing to the layout in any reader.
What am I getting wrong or miss understanding please?
来源：https://stackoverflow.com/questions/26426345/itext-correctly-get-each-image-position-within-page-or-document