1) Open the attached file "a.pdf"
2) Push "Scan & OCR"
3) Push "Correct recognized text."
The text window should look like the attached image "1.jpg". The 3 areas of recognized text are erroneous recognitions. One is a square angle denoting a corner of a page, the other two are dirty spots. The only thing I can do now is replace the recognized text with a space (" "). I suggest that a better option would be to add another button to the dialog box shown in the attached file "2.jpg." The button could be called "Remove recognition" or something along those lines. This would be the best way to deal with erroneous recognitions I believe.