Is there a way to make Amazon Textract return the skew angle of the pdf document that it is processing?
When I do detection on a document that has been scanned it, it attempts to de-skew it but it's calculations are slightly off. There doesn't seem to be a way that I can find to access the raw bounding box or geometry of each LINE or WORD, it is all fixed to de-skewed position.
Can I find the skew angle another way or can I disable de-skewing on the request somehow?