[PATCH 09/10] ipa: rkisp1: awb: Take the CCM into account for the AWB gains calculation

[Kieran Bingham]

Quoting Stefan Klug (2025-02-17 10:01:50)
> Apparently AWB measurements are taken after the CCM. So the estimated

Are we more sure than 'apparantly' ? What do we need to do to be more
confident or certain of this ?

> colour temperature and the corresponding CCM also lead to changed
> rgbMeans. This is another source of oscillations. Fix that by applying
> the inverse transform on the rgbMeans.
> 
> Signed-off-by: Stefan Klug <stefan.klug at ideasonboard.com>
> ---
>  src/ipa/rkisp1/algorithms/awb.cpp | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/src/ipa/rkisp1/algorithms/awb.cpp b/src/ipa/rkisp1/algorithms/awb.cpp
> index 12934771c69c..66e6aecedc4c 100644
> --- a/src/ipa/rkisp1/algorithms/awb.cpp
> +++ b/src/ipa/rkisp1/algorithms/awb.cpp
> @@ -305,6 +305,8 @@ void Awb::process(IPAContext &context,
>                 rgbMeans = rgbMeans.max(0.0);
>         }
>  

I would definitely put a block comment here describing /why/ we are
doing this - likely matching closely to what you have in the commit
message.

> +       rgbMeans = frameContext.ccm.ccm.cast<double>().inverse() * rgbMeans;
> +
>         /*
>          * The ISP computes the AWB means after applying the colour gains,
>          * divide by the gains that were used to get the raw means from the
> -- 
> 2.43.0
>