Text being extracted as an image

In one of the examples for the PDF Extract API, some of the text is included in a png along with the adjacent image. The text is not present in the structuredData.json file, so it is essentially not being extracted.

Is this a common issue with the Extract API?