Is there an option to let pdfsandwich try 90° rotations automatically for scanned pages when necessary?

I am on Ubuntu. Most of my scanned documents are German, English or French. Some scans have to be rotated before doing OCR on them, otherwise pdfsandwich returns nonsense OCR. Is there any preprocessor parameter which lets pdfsandwich automatically try rotated input files if the OCR result from tesseract from an unrotated input file is of low quality? Is there any preprocessing which first looks for the general orientation of lines ad which rotates them to support better OCR?