From 10416606df247aba50fab6b0ea25f37424e90bd0 Mon Sep 17 00:00:00 2001 From: Jeffery Tang Date: Tue, 3 Sep 2019 11:35:05 -0400 Subject: [PATCH] Typo --- doc/HWSetup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/HWSetup.md b/doc/HWSetup.md index efa8212..43e5922 100644 --- a/doc/HWSetup.md +++ b/doc/HWSetup.md @@ -61,7 +61,7 @@ It is then simply replayed on a layer under the UI with a transparent background ## Motion detection The motion detection is done using the motion vectors that raspivid can output to a separate file for each video chunk. -These files are read in and each frame we chck if there are enough motion vectors big enough to consider it movement. +These files are read in and each frame we check if there are enough motion vectors big enough to consider it movement. To avoid false positives due to camera vibration we require a number of contiguous frames with movement. All config related to motion detection depends a lot on the parameters of the camera (resolution, chunk size, etc...) and might not work correctly if the config changes. You can find more info here: https://www.raspberrypi.org/blog/vectors-from-coarse-motion-estimation/