If you want to help wyze with the new detection mode, check this post:
2 Likes
All: After a month of hoping for an improved firmware, I decided to experiment with the new detection zone feature. I finally have my false positives to a minimum.
First: the coloring of the zone tiles is confusing. The translucent gray tiles are those which are excluded from detection and the transparent tiles are included. I went back and forth on this a few times before I figured it out (thinking it was the opposite)
Second: I started with a single active tile in my detection zone – one where I knew there would be no false movement detected. Once I saw the event count go to zero, I expanded the size of my tiled area being very conservative on the size of my detection zone. If I started getting false positives, I used the green motion tagging rectangle to see which tiles to exclude.
After fiddling with this and thinking about it, I filed two wishlist topics:
First, a request that motion-tagging NOT compute a rectangle based on ALL the tiles which had motion, but it should compute a non-rectanglular (tile based) outline. This way, a diagonal row of bushes doesn’t cause a huge motion tagging rectangle simply becuase one leaf on each end of the row moved.
Second: a request for dynamically-computed detection zones which keep track of over-active regions and automatically excludes those from detection. A time-decayed list of active tiles could be maintained by the firmware and the static user-defined detection zone would not be needed (or could be optional as a fixed exclusion zone say for a hot tub where intimate things might occur, nudge nudge, wink, wink). In addition to solving the blowing leaf problem, this would also handle earth’s rotation and shadows being cast in different places as the day progresses. A static detection zone simply cannot deal with this without excluding far more than necessary.