luisomoreau

Hello Dennis,Thank you for your feedback. Yes I was getting some errors as well. You need to train your model with different brightness, angles, etc... to have better results.I tried this tutorial with 7-9 people and I got a 80% of success in average.

Hello Dennis,Unfortunately, only the blue dots have coverage. We are under very active network deployment in the US.Best regards,

Hello!!Where are you based in the US?You can checkout if you have Sigfox coverage in your area: https://www.sigfox.com/en/coverageAnd you can buy an equivalent that will work in the US on: https://seasluglabs.io/collections/frontpage/produ...Best regards,Louis

Hello!It is here: https://github.com/luisomoreau/sigfox_face_recogni...Regards,

Hello fatty_inc,You can use either an ZigBee shield with a usb receiver or a BLE shield with a USB receiver (needed to program the device over the air). I thought it was included in the air board package but you are right, it is not included.

What is the value of the GPS timeout have you tried to increase it up to 120 seconds (it should be more than enough)?

Try outside you should have better results ;)

Are you testing inside a building?

Hi Jose MaríaS4, Are you sure you set the geolocalisation "default" (for my tutorial) or the "Telecom Design" as the display type in Sigfox Backend?Go to "Device Type" -> "Information" and click on edit.By the way I'm developing an open source app to decode the GPS data (and other data type) https://github.com/luisomoreau/iot-platformThere is no documentation yet but I'm working on it.

Great tutorial Antoine!I really appreciate the 3D printing casing bonus.Thanks

Hi!Apparently Avnet is the exclusive distributor...You can contact Telecom Design directly, they are pretty supportive in general.contact@td-next.com

This one?https://www.instructables.com/file/F0RXPPRIRLQG01S/If so, yes :)

Sorry for the slow reply...

Hi senaug82,As far as I remember, I have just use the timer to wake the device up...This project was not power efficient at all ;)

Hi Kallell!As far as I remember (I haven't played with this project for few month now), it really depended on the lightening conditions.You can try to adjust the value of the scale factor from 1.3 to 1.2faces = faceCascade.detectMultiScale(gray,scaleFactor=1.3,minNeighbors=5,minSize=(30, 30),flags=cv2.cv.CV_HAAR_SCALE_IMAGEConcerning the confidence values, try with a value between 2000 and 3000. It also depends on how many pictures you took and at what conditions (smiling, very bright, very dark, turning your head right and left, etc...)I hope this will help :)

Hi! I know there is a quick trick to do for the RPI3 in the rapsbian config to allow the camera...Did you have a look at this link? http://www.pyimagesearch.com/2016/04/18/install-guide-raspberry-pi-3-raspbian-jessie-opencv-3/Let me know if it worked :)

Hi gideonBreo, The cable between the EFM32 and my laptop is a miniUSB cable. The cable between the EVB and my laptop is an FTDI cable. I hope this helps :)

Here is a picture if it is needed ;)

Hi Stan1y,Subscriptions to the Sigfox service are free for developers (developement kits & evaluation boards). Otherwise, they range between 1€/month & 1€/year depending of the volume commitment.Regarding coverage: the active coverage map is publicly available on http://sigfox.com/coverage I hope this will help :)