Skip to content

CVE-2026-73067 Tesseract: Heap OOB read in the DAWG loader

MediumCVSS 6.7 EPSS 0.13%
Published
Updated
Exploit status
No public exploit known

Description

Tesseract is an open source OCR engine. Prior to 5.5.3, a crafted .traineddata model loaded through TessBaseAPI::Init can cause SquishedDawg::read_squished_dawg in src/dict/dawg.cpp to accept an unterminated forward-edge run, after which SquishedDawg::Load calls num_forward_edges(0) and last_edge in src/dict/dawg.h reads beyond edges_, causing a heap out-of-bounds read and process crash before image processing. This issue is fixed in version 5.5.3.

Affected products

ProductAffected versionsFixed in
tesseract-ocr tesseract = < 5.5.3 No fix listed
ubuntu/tesseract(Ubuntu:22.04) < 4.1.1-2.1ubuntu0.1~esm1 4.1.1-2.1ubuntu0.1~esm1
ubuntu/tesseract(Ubuntu:24.04) < 5.3.4-1ubuntu0.1~esm1 5.3.4-1ubuntu0.1~esm1
ubuntu/tesseract(Ubuntu:26.04) < 5.5.0-1ubuntu0.1~esm1 5.5.0-1ubuntu0.1~esm1
ubuntu/tesseract(Ubuntu:26.10) All versions No fix listed
debian/tesseract(Debian:12) All versions No fix listed
debian/tesseract(Debian:14) < 5.5.4-1 5.5.4-1
debian/tesseract(Debian:13) All versions No fix listed
debian/tesseract(Debian:12) = 5.3.0-2 No fix listed
debian/tesseract(Debian:12) = 5.3.3-1 No fix listed
debian/tesseract(Debian:12) = 5.3.4-1 No fix listed
debian/tesseract(Debian:12) = 5.3.4-1.1 No fix listed
debian/tesseract(Debian:12) = 5.3.4-1.2 No fix listed
debian/tesseract(Debian:12) = 5.3.4-1.3 No fix listed
debian/tesseract(Debian:12) = 5.3.4-1.4 No fix listed
debian/tesseract(Debian:12) = 5.5.0-1 No fix listed
debian/tesseract(Debian:12) = 5.5.0-1.1 No fix listed
debian/tesseract(Debian:12) = 5.5.4-1 No fix listed
debian/tesseract(Debian:13) = 5.5.0-1 No fix listed
debian/tesseract(Debian:13) = 5.5.0-1.1 No fix listed
debian/tesseract(Debian:13) = 5.5.4-1 No fix listed
debian/tesseract(Debian:14) = 5.5.0-1 No fix listed
debian/tesseract(Debian:14) = 5.5.0-1.1 No fix listed

References