[PATCH 1/2] utils: raspberrypi: ctt: Fix NaNs in lens shading tables

David Plowman david.plowman at raspberrypi.com
Mon Apr 28 12:36:03 CEST 2025


The problem occurs when the calculation could lead to a final row (or
column) of grid squares with no pixels in them (and hence, NaNs).

One specific case is a Pi 5 with an image width (or height) of 1364,
so that's 682 Bayer quads. To give 32 grid squares it was calculating
22 quads per cell. However, 31 * 22 = 682 leaving nothing in the final
column.

The fix is to do a rounding-down division by the number of cells minus
one, rather than a rounding-up division by the number of cells. This
turns the corner case from one where the final row/column has no
pixels to one where we don't quite cover the full image, which is how
we have to handle these cases.

Signed-off-by: David Plowman <david.plowman at raspberrypi.com>
---
 utils/raspberrypi/ctt/ctt_alsc.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/utils/raspberrypi/ctt/ctt_alsc.py b/utils/raspberrypi/ctt/ctt_alsc.py
index 1d94dfa5..f4fd09e3 100644
--- a/utils/raspberrypi/ctt/ctt_alsc.py
+++ b/utils/raspberrypi/ctt/ctt_alsc.py
@@ -127,11 +127,12 @@ def alsc(Cam, Img, do_alsc_colour, plot=False, grid_size=(16, 12), max_gain=8.0)
     channels = [Img.channels[i] for i in Img.order]
     """
     calculate size of single rectangle.
-    -(-(w-1)//32) is a ceiling division. w-1 is to deal robustly with the case
-    where w is a multiple of 32.
+    The divisions here must ensure the final row/column of cells has a non-zero number of
+    pixels.
     """
     w, h = Img.w/2, Img.h/2
-    dx, dy = int(-(-(w-1)//grid_w)), int(-(-(h-1)//grid_h))
+    dx, dy = (w - 1) // (grid_w - 1), (h - 1) // (grid_h - 1)
+
     """
     average the green channels into one
     """
-- 
2.34.1



More information about the libcamera-devel mailing list