Impact
The vulnerability involves an out‑of‑bounds heap write in the GenericVector::read function of the Tesseract OCR engine. It occurs when the function reads two independent integer fields, reserved and size_used_, from a .traineddata file without validating that the allocated buffer is large enough for size_used a backing array based on the reserved value but proceeds to write size_used_ FontInfo structures, overrunning the buffer and corrupting adjacent heap objects. This can cause a crash or, if exploited, controlled memory corruption and is a classic heap buffer overflow (CWE‑787).
Affected Systems
The affected product is the Tesseract OCR engine, version 5.5.3 and all earlier releases. No newer releases have a fix at the time of this review.
Risk and Exploitability
The CVSS base score of 8.6 indicates high severity. The vulnerability requires a crafted .traineddata file with a version_id of 4 or later. Attackers are most likely to supply a malicious model to an OCR service that loads user files, such as a web‑based OCR API or and the vulnerability is not listed in CISA KEV, but the lack of a patch and the potential for heap corruption make the risk significant.
OpenCVE Enrichment