Found scandata at /var/tmp/autoclean/derive/annualreport201011harv/annualreport201011harv_scandata.xml.
Will skip page 1 because it's a cover page.
Will skip page 36 because it's a cover page.
Seeking title page...

Will skip the following pages as requested: 1,36
Title:      "annual report"
Start word: "annual"

Candidate words:
     p. 3     report.             41 px (0.14)
     p. 4     reports             43 px (0.14)

Selected title page: 4
Identified leaf 4 as title page.
Marking leaf 4 as 'Title'.
Scandata has been modified. Saving log...