← 76834d64-8bb7-463e-a952-42280d97f2bc_September-LLB-Statistics_website-artifacted.pdf
OCR Overlay — PaddleOCR-VL (Baidu)
Boxes are the OCR items for each page, colored and labeled by field type (numbering restarts each page). view JSON
Page 1
6 item(s)
CAPTION (1)
FOOTNOTE (1)
PAGE_FOOTER (1)
PICTURE (1)
SECTION_HEADER (1)
TEXT (1)
1 · SECTION_HEADER
2 · CAPTION
3 · PICTURE
4 · FOOTNOTE
5 · TEXT
6 · PAGE_FOOTER
Page 2
3 item(s)
FOOTNOTE (2)
TABLE (1)
1 · FOOTNOTE
2 · FOOTNOTE
3 · TABLE
Page 3
7 item(s)
FOOTNOTE (4)
PICTURE (1)
TEXT (2)
1 · PICTURE
2 · FOOTNOTE
3 · FOOTNOTE
4 · FOOTNOTE
5 · FOOTNOTE
6 · TEXT
7 · TEXT
Page 4
2 item(s)
CAPTION (1)
TABLE (1)
1 · CAPTION
2 · TABLE
Page 5
4 item(s)
CAPTION (2)
PICTURE (1)
TABLE (1)
1 · CAPTION
2 · CAPTION
3 · TABLE
4 · PICTURE