The Image-Extraction Option "When possible, use images internal file format" is great to get unmodified jpgs out of the PDF.
It would be much more helpfull, if it detects G4-TIFFs too. (FilterType=="CCITTFaxDecode"), this would improve output size damatically.
Example:
JAZZFAKE.PDF has 40.185.734 Bytes
Exported
=== AS-BMP === (will give the tenfold size )
2 .jpg 1.239.812 bytes
446 .bmp 471.289.092 bytes
=== AS-PNG === (is already quite good, but still not the original files)
2 .jpg 1.239.812 bytes
446 .png 43.312.693 bytes
=== AS-TIFF === (with my own script, quite unbeatable )
2 .jpg 1.239.807 bytes
446 .tif 21.874.766 bytesI will test this gladly, if you implement it.