Confidence Threshold
Cutoff score used to accept or reject OCR-detected characters/words.
Definition
Cutoff score used to accept or reject OCR-detected characters/words.
Why it matters
Balances recall and precision in text extraction outcomes.
Related PDFHQ workflows
Relevant for extraction quality tuning in OCR-heavy workflows.
Related Knowledge
OCR Confidence
Score indicating OCR engine certainty in recognized text.
Text Layer
Machine-readable text content embedded alongside scanned page imagery.
Thresholding
Pixel-level decision method used in binarization to separate foreground text from background.
Table Extraction
Process of detecting and exporting table structures from documents.