Originally Posted by eugentango
I did not understand what exactly I have to do in Calibre. Or you mean I have to edit the PDF file?
I tried to look for code in ePub file - there is a possibilite in calibre to open it in ePub-viewer and see the html-code. But I did not found any possibility to flip it back.
Could you describe the process more precise? For dummies...
It is NOT
the HTML or CSS code in the ePub that is your problem so don't go looking for them or for ways to edit them.
It is, as frostschutz stated, an issue with the images themselves. There as some types of images at can exist in a PDF that calibre does not handle. Sometimes they get "lost" and sometimes that get misprocessed. The latter is what happened in this case.
You need to extract the images from the ePub and use an image editor to fix the problems (flip, rotate, invert, whatever...) and then copy the fixed images back into the ePub.
You can use calibre's Tweak Epub function to expand the ePub to get access to the images. You can also change the ePub's extension to ZIP, extract copies of the images, edit the images, and then copy the fixed images back into the ZIP before changiing the extension back to EPUB.