Bug 1752025 (CVE-2019-14491)

Summary: CVE-2019-14491 opencv: out-of-bounds read in function cv::predictOrdered<cv::HaarEvaluator> in modules/objdetect/src/cascadedetect.hpp leads to dos
Product: [Other] Security Response Reporter: Dhananjay Arunesh <darunesh>
Component: vulnerabilityAssignee: Red Hat Product Security <security-response-team>
Status: CLOSED CURRENTRELEASE QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: unspecifiedCC: andrew, databases-maint, hhorak, jkucera, jmlich83, jridky, karlthered, kwizart, pkajaba, rakesh.pandit, viktor.vix.jancik
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: opencv 3.4.7, opencv 4.1.2 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2020-07-28 08:10:25 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 1752026, 1774119, 1774120    
Bug Blocks: 1752029    

Description Dhananjay Arunesh 2019-09-13 13:24:58 UTC 
An issue was discovered in OpenCV before 3.4.7 and 4.x before 4.1.1. There is an out of bounds read in the function cv::predictOrdered<cv::HaarEvaluator> in modules/objdetect/src/cascadedetect.hpp, which leads to denial of service.

Reference:
https://github.com/opencv/opencv/compare/371bba8...ddbd10c
https://github.com/opencv/opencv/issues/15125
https://github.com/opencv/opencv/compare/33b765d...4a7ca5a
Comment 1 Dhananjay Arunesh 2019-09-13 13:25:38 UTC 
Created opencv tracking bugs for this issue:

Affects: fedora-all [bug 1752026]
Comment 3 Stefan Cornelius 2019-11-19 16:15:31 UTC 
Patch:
https://github.com/opencv/opencv/commit/321c74ccd6077bdea1d47450ca4fe955cb5b6330
Comment 6 Stefan Cornelius 2019-11-19 16:36:09 UTC 
Statement:

It's possible to cause opencv to attempt to read from incorrect or invalid memory when loading specially crafted classifiers (trained data used for object detection), possibly leading to a crash. Although it's technically possible that classifiers are used from untrusted sources, it's  probably an unlikely case in practice.
Comment 7 Sergio Basto 2019-11-20 02:06:01 UTC 
(In reply to Stefan Cornelius)

> Fixed In Version: opencv 3.4.7, opencv 4.1.2

Since in master we already got 4.1 , I will update F31 with 3.4.7
Comment 8 Nicolas Chauvet (kwizart) 2020-07-28 08:10:25 UTC 
Fixed in 3.4.10-1 package in f31