Date
Attendees
Goals
Discussion items
| Time | Item | Who | Notes |
|---|---|---|---|
| OCR | Liz | OCR not working well, even with very text-based images. | |
| Paul | 171k images. 86k have tiff file "largest." 84k have tiff file called "larger." Difference in resolution. Jeff: need to develop ranges to process | ||
| Jeff | Would be nice to know how long it would take to run full corpus to make case for using supercomputers |
Action items
- Sandeep Puthanveetil Satheesan: Can we improve the OCR through thresholding or anything else to enhance?
- Paul Rodriguez: Investigating different resolutions sizes based on file name "large" and "larger," etc.