Path: news.eternal-september.org!eternal-september.org!feeder3.eternal-september.org!border-4.nntp.ord.giganews.com!nntp.giganews.com!Xl.tags.giganews.com!local-4.nntp.ord.giganews.com!news.giganews.com.POSTED!not-for-mail
NNTP-Posting-Date: Sun, 04 May 2025 05:24:12 +0000
Newsgroups: comp.os.linux.misc
X-Mozilla-News-Host: snews://news.giganews.com:563
From: c186282 <c186282@nnada.net>
Subject: Advice on "Matrix Metering" for Cams ?
Date: Sun, 4 May 2025 01:24:21 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.13.0
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Message-ID: <uhucnXb6KohhZIv1nZ2dnZfqn_GdnZ2d@giganews.com>
Lines: 44
X-Usenet-Provider: http://www.giganews.com
X-Trace: sv3-7VRxQVvhvbxsJ+XQLb0I72utCINIiH7cIfdNtLFiwZUHm5bQG4E1mQ/nnQQ4vAjXfzZbGzY5DoescOR!9eGbSr+wVz3g/b7i57iOw+urEuPbLGvLO/V5ziJqkHLx6oDG1AddQrWhgt1qws4f23csxWy16YPd
X-Complaints-To: abuse@giganews.com
X-DMCA-Notifications: http://www.giganews.com/info/dmca.html
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
X-Postfilter: 1.3.40

I've got half a dozen security cams. Some are
straight IP, others - PI and others - use 'motion'
to create an MPEG stream.

Wrote a display app (python) that rolls through
the cams so I can see what's going on. Also a
sep app that RECORDS selected cams as MP4s
mostly using OpenCV stuff.

HOWEVER ... while I do have a little daemon that
sets day/night values for v4l2 that's not always
good enough. Super-bright, or cloudy, days can
result in kinda crappy images.

Tried simple histogram expansion - but that kinda
looked WORSE in most cases - TOO much contrast.

SO ... wrote a little app that breaks the images
into a 7x5 matrix and yields 'average brightness'
figures for each segment ... the basic of 'matrix
metering'.

Alas even after extensive searching I can't find
any advice on HOW to make use of those figures,
HOW to adjust brightness/contrast based on that
matrix analysis.

Camera makers HAVE figured this out, and 'matrix'
almost always yields great images. Alas the HOW
is not published.

OpenCV2 can adjust both brightness and contrast
quite easily - but what VALUES to feed it .....

Have you seen anybody addressing this problem,
this issue, on the net where I might get advice
on how best to analyze those image segments and
derive the needed adjustment figures ??? This
drifts into "statistics" - not my best issue.