Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Re: Motion sensor that can "see" through glass
108 views
Skip to first unread message
Axel Berger
Dec 23, 2021, 5:03:28 PM12/23/21
to
Chris Schram wrote:
> It would be used in some sort of RasPi or Arduino configuration.
That might work, but it's a lot of hassle and complication. The doorbell
already is an electric signal. Could you not use that directly?
--
/¯\ No | Dipl.-Ing. F. Axel Berger Tel: +49/ 221/ 7771 8067
\ / HTML | Roald-Amundsen-Straße 2a Fax: +49/ 221/ 7771 8069
X in | D-50829 Köln-Ossendorf http://berger-odenthal.de
/ \ Mail | -- No unannounced, large, binary attachments, please! --
> It would be used in some sort of RasPi or Arduino configuration.
That might work, but it's a lot of hassle and complication. The doorbell
already is an electric signal. Could you not use that directly?
--
/¯\ No | Dipl.-Ing. F. Axel Berger Tel: +49/ 221/ 7771 8067
\ / HTML | Roald-Amundsen-Straße 2a Fax: +49/ 221/ 7771 8069
X in | D-50829 Köln-Ossendorf http://berger-odenthal.de
/ \ Mail | -- No unannounced, large, binary attachments, please! --
David Higton
Dec 23, 2021, 5:25:31 PM12/23/21
to
In message <61C4F25D...@Berger-Odenthal.De>
the hall in conventional use, the other is hooked up to an old BeagleBoard
that had become spare. I found that the electronics could be powered
from 5 volts, available from the BB, so it didn't need plugging in to
the mains - fortunate, as the power supply is a non-isolated type.
I wrote some software to send a message to Pushover, which sends me an
alert on my mobile phone. It usually arrives within 2 seconds, but it
has to be admitted that occasionally it takes much longer, so it's
definitely not ideal.
The same idea can be used with a conventional doorbell. It's just a
case of appropriate interfacing.
As for why: in our house, if we're in the sun lounge, the ringer in
the hall is completely inaudible. Same applies if we're in the garden.
There are other places where we can be, and the sound of what we're
doing can drown the doorbell. So it's a different use case, but it
boils down to the same thing.
If anyone knows a way to get notifications to a mobile phone that are
always timely, I'd like to know.
David
Axel Berger <Sp...@Berger-Odenthal.De> wrote:
> Chris Schram wrote:
> > It would be used in some sort of RasPi or Arduino configuration.
>
> That might work, but it's a lot of hassle and complication. The doorbell
> already is an electric signal. Could you not use that directly?
I did that. I bought a wireless doorbell with two "ringers"; one is in
> Chris Schram wrote:
> > It would be used in some sort of RasPi or Arduino configuration.
>
> That might work, but it's a lot of hassle and complication. The doorbell
> already is an electric signal. Could you not use that directly?
the hall in conventional use, the other is hooked up to an old BeagleBoard
that had become spare. I found that the electronics could be powered
from 5 volts, available from the BB, so it didn't need plugging in to
the mains - fortunate, as the power supply is a non-isolated type.
I wrote some software to send a message to Pushover, which sends me an
alert on my mobile phone. It usually arrives within 2 seconds, but it
has to be admitted that occasionally it takes much longer, so it's
definitely not ideal.
The same idea can be used with a conventional doorbell. It's just a
case of appropriate interfacing.
As for why: in our house, if we're in the sun lounge, the ringer in
the hall is completely inaudible. Same applies if we're in the garden.
There are other places where we can be, and the sound of what we're
doing can drown the doorbell. So it's a different use case, but it
boils down to the same thing.
If anyone knows a way to get notifications to a mobile phone that are
always timely, I'd like to know.
David
Lew Pitcher
Dec 23, 2021, 5:29:07 PM12/23/21
to
On Thu, 23 Dec 2021 21:53:27 +0000, Chris Schram wrote:
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
I have purchased (through AliExpress) a number of RCWL0516 "Doppler Microwave"
motion sensors that provide (among other things) a single-bit binary interface
suitable for use with one of the Raspberry PI's digital pins.
As the device uses weak microwave radiation, it can sense motion through glass,
brick or drywall, and has an adjustable range of up to 5 metres (about 15 feet)
depending on the intervening media.
These devices typically cost less that $1.00 (excluding shipping).
Perhaps one of these would work for you.
--
Lew Pitcher
"In Skills, We Trust"
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
I have purchased (through AliExpress) a number of RCWL0516 "Doppler Microwave"
motion sensors that provide (among other things) a single-bit binary interface
suitable for use with one of the Raspberry PI's digital pins.
As the device uses weak microwave radiation, it can sense motion through glass,
brick or drywall, and has an adjustable range of up to 5 metres (about 15 feet)
depending on the intervening media.
These devices typically cost less that $1.00 (excluding shipping).
Perhaps one of these would work for you.
--
Lew Pitcher
"In Skills, We Trust"
Chris Schram
Dec 23, 2021, 6:40:59 PM12/23/21
to
On 2021-12-23, Axel Berger <Sp...@Berger-Odenthal.De> wrote:
> Chris Schram wrote:
>> It would be used in some sort of RasPi or Arduino configuration.
>
> That might work, but it's a lot of hassle and complication. The doorbell
> already is an electric signal. Could you not use that directly?
>
Heh! You don't know about my doorbell. It's a brass replica of an
> Chris Schram wrote:
>> It would be used in some sort of RasPi or Arduino configuration.
>
> That might work, but it's a lot of hassle and complication. The doorbell
> already is an electric signal. Could you not use that directly?
>
antique. You twist a knob on the outside of the door which spins a
clapper against the bell on the inside. No electronics, or even
electricity involved.
--
chri...@me.com is a filtered spam magnet. Email replies may be lost.
You're better off replying to this newsgroup.
Chris Schram
Dec 23, 2021, 6:48:01 PM12/23/21
to
also occurs to me that there may be some Arduino peripherals that can
output some relatively loud noise. I may be able to forego the Pi
altogether. Thanks for the quick response.
Axel Berger
Dec 23, 2021, 9:03:09 PM12/23/21
to
Chris Schram wrote:
> Heh! You don't know about my doorbell.
Correct. In that case I modify my suggestion and suggest that a
> Heh! You don't know about my doorbell.
microphone with some very simple electronics right next to the clanger
is simpler and more relible than anything optical.
GlowingBlueMist
Dec 24, 2021, 12:38:48 AM12/24/21
to
On 12/23/2021 3:53 PM, Chris Schram wrote:
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
You might want to look into the doorbell units sold at this link.
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
https://www.amazon.com/Wireless-Doorbell-1PlusOne-Waterproof-Transmitter/dp/B07C24N61X/ref=psdc_6291370011_t2_B08XYKY9R2
The claim you can purchase more receivers or units and using push-button
settings you can connect them all together and have the doorbell ring in
say 8 rooms or how ever many you combine together.
No, I have not used the product myself.
Andy Burns
Dec 24, 2021, 2:33:59 AM12/24/21
to
Chris Schram wrote:
> I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
Look at Andreas "guy with the swiss accent" Spiess on youtube, he has videos on
> I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
various sensor types, including radar type, and connecting to rPi and
microcontrollers.
e.g. this playlist, but search his channel for more ...
<https://www.youtube.com/playlist?list=PL3XBzmAj53RmryaFf7vyAP6U3f7H_5Lfk>
The Natural Philosopher
Dec 24, 2021, 3:39:57 AM12/24/21
to
On 23/12/2021 21:53, Chris Schram wrote:
> I have grown more and more hard of hearing as I grow older. and I cannot
--
“But what a weak barrier is truth when it stands in the way of an
hypothesis!”
Mary Wollstonecraft
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
I would be inclined to think in terms of burglar alarm sensors.
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
--
“But what a weak barrier is truth when it stands in the way of an
hypothesis!”
Mary Wollstonecraft
Chris Green
Dec 24, 2021, 4:03:04 AM12/24/21
to
motion sensors work through doors, windows, etc.
Over here (UK) the DIY sheds have both PIR and microwave sensors available
quite cheaply.
--
Chris Green
·
A. Dumas
Dec 24, 2021, 4:40:34 AM12/24/21
to
Axel Berger <Sp...@Berger-Odenthal.De> wrote:
> Chris Schram wrote:
>> Heh! You don't know about my doorbell.
>
> Correct. In that case I modify my suggestion and suggest that a
> microphone with some very simple electronics right next to the clanger
> is simpler and more relible than anything optical.
Easy to set up initially but might be very hard (fft) if it has to
> Chris Schram wrote:
>> Heh! You don't know about my doorbell.
>
> Correct. In that case I modify my suggestion and suggest that a
> microphone with some very simple electronics right next to the clanger
> is simpler and more relible than anything optical.
distinguish the bell from other sounds, if those are about the same level.
So I'd be inclined to fit a microswitch to the bell mechanism. Easy, direct
on/off signal but probably a little more finicky to set up.
Axel Berger
Dec 24, 2021, 6:37:55 AM12/24/21
to
"A. Dumas" wrote:
> Easy to set up initially but might be very hard (fft) if it has to
> distinguish the bell from other sounds, if those are about the same level.
> So I'd be inclined to fit a microswitch to the bell mechanism. Easy, direct
> on/off signal but probably a little more finicky to set up.
I disagree about the reliability. Sound decays with the square root law.
> Easy to set up initially but might be very hard (fft) if it has to
> distinguish the bell from other sounds, if those are about the same level.
> So I'd be inclined to fit a microswitch to the bell mechanism. Easy, direct
> on/off signal but probably a little more finicky to set up.
If you place a microphone right nexr to the source it will be hard to
confuse it with an external equally strong signal anf if that happens it
should be obvious.
You can get a mechanical switch to be reliable for a long tinme, but
it's hard.
The suggested optical (or microwave) slution will be the least reliable
of all, I expect.
Chris Elvidge
Dec 24, 2021, 8:02:12 AM12/24/21
to
On 23/12/2021 21:53, Chris Schram wrote:
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
Assuming you already have a "push" doorbell on the outside, can you not
> I have grown more and more hard of hearing as I grow older. and I cannot
> always hear when someone is at the front door. I'm looking for a motion
> sensor that I can locate inside the front window.
>
> I have played briefly with an old Radio Shack infrared motion alarm that
> otherwise works fine, but can't "see" through glass at all.
>
> I also have a little Wyze security camera that can detect motion, but it
> seems way TOO sensitive to be useful.
>
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
>
attach a lightbulb in parallel with the inside ringer?
Alternatively
https://www.amazon.co.uk/gp/product/B01LQBRKXG/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1
It has a blue flashing light on the receiver.
--
Chris Elvidge
England
Folderol
Dec 24, 2021, 10:05:45 AM12/24/21
to
On Thu, 23 Dec 2021 21:53:27 -0000 (UTC)
Chris Schram <chri...@me.com> wrote:
>I have grown more and more hard of hearing as I grow older. and I cannot
>always hear when someone is at the front door. I'm looking for a motion
>sensor that I can locate inside the front window.
>
>I have played briefly with an old Radio Shack infrared motion alarm that
>otherwise works fine, but can't "see" through glass at all.
>
>I also have a little Wyze security camera that can detect motion, but it
>seems way TOO sensitive to be useful.
>
>Could ultrasonic be the way to go? That's a technology I've yet to play
>with. It would be used in some sort of RasPi or Arduino configuration.
>
Use a short range proximity photocell pointing at the clapper mechanism,
Chris Schram <chri...@me.com> wrote:
>I have grown more and more hard of hearing as I grow older. and I cannot
>always hear when someone is at the front door. I'm looking for a motion
>sensor that I can locate inside the front window.
>
>I have played briefly with an old Radio Shack infrared motion alarm that
>otherwise works fine, but can't "see" through glass at all.
>
>I also have a little Wyze security camera that can detect motion, but it
>seems way TOO sensitive to be useful.
>
>Could ultrasonic be the way to go? That's a technology I've yet to play
>with. It would be used in some sort of RasPi or Arduino configuration.
>
then an arduino to monitor *change* of state, so it doesn't matter if the mech
stops with the photocell obscured or not.
Something like this:
https://www.amazon.co.uk/Photocell-Infrared-Photoelectric-Proximity-Outside/dp/B093HGNWT1/ref=sr_1_11?_encoding=UTF8&c=ts&keywords=Photoelectric+Sensors&qid=1640358044&s=industrial&sr=1-11&ts_id=10256556031
Use two of the arduino outputs in antiphase at (say) 2kHz to directly drive a
ceramic sounder.
Done this sort of thing loads of times :)
--
Basic
druck
Dec 24, 2021, 4:28:15 PM12/24/21
to
On 23/12/2021 21:53, Chris Schram wrote:
> Could ultrasonic be the way to go? That's a technology I've yet to play
> with. It would be used in some sort of RasPi or Arduino configuration.
Ultrasonic wont see through glass or the clear lid of an electronic box,
> with. It would be used in some sort of RasPi or Arduino configuration.
(I did check to be sure). So it would need to be mounted externally, and
then you have the problem of weather proofing.
---druck
Computer Nerd Kev
Dec 24, 2021, 4:48:20 PM12/24/21
to
water level sensor and the sensor is completely sealed up. Then again
the same applies to IR solutions, so one has to assume the OP just
doesn't like electronics outside.
I didn't see anyone mention mounting a laser pointing at a
slightly-angled mirror outside which reflects the beam back to a
light sensor. Just one more option, so long as it can be mounted
at a height where people won't accidentally look into it.
--
__ __
#_ < |\| |< _#
Chris Elvidge
Dec 24, 2021, 5:19:47 PM12/24/21
to
On 24/12/2021 16:32, Dennis Lee Bieber wrote:
> On Thu, 23 Dec 2021 21:53:27 -0000 (UTC), Chris Schram <chri...@me.com>
> declaimed the following:
> warm when the sun shines on glass. Ultraviolet light passes through glass,
> gets absorbed by fabrics and what-have-you on the inside, and those items
> then re-radiate the energy in the infrared band.
> pretty much have to fit the sensor (for almost anything except image
> processing type detection [this image differs from last image]) outside the
> window and run leads to the R-Pi (or whatever you use to process data).
>
> A PIR sensor /might/ be usable in daylight (they are commonly used for
> auto-activation motion detect security lights). They trigger on an IR
> source moving between segments in the sensor window.
> https://learn.adafruit.com/pir-passive-infrared-proximity-motion-sensor/how-pirs-work
>
> Ultrasonic ("PING)))" -- yes, the ))) is part of the name for some)
> sensors are more distance detection items than motion sensors, and may not
> have a great range. Some old Polaroid cameras used them for focusing -- but
> one has to account for the default focus being "infinity", and
> depth-of-field on those being deep enough that anything around 15+ feet was
> considered "infinity".
>
> https://www.parallax.com/product/ping-ultrasonic-distance-sensor/ (10ft
> range)
> https://learn.adafruit.com/ultrasonic-sonar-distance-sensors (suggests
> 13ft, but recommends 8ft)
>
>
I've got a Raspberry Pi B with a Pi Camera mounted and use the 'motion'
package to record video when motion is detected - through a window.
Video is not very high quality, but you don't need that.
You could use inotify to watch the destination directory and email you
when motion is recorded.
--
Chris Elvidge
England
> On Thu, 23 Dec 2021 21:53:27 -0000 (UTC), Chris Schram <chri...@me.com>
> declaimed the following:
>
>>
>> I have played briefly with an old Radio Shack infrared motion alarm that
>> otherwise works fine, but can't "see" through glass at all.
>>
> Not really surprising -- it's one reason car and home interiors get
>>
>> I have played briefly with an old Radio Shack infrared motion alarm that
>> otherwise works fine, but can't "see" through glass at all.
>>
> warm when the sun shines on glass. Ultraviolet light passes through glass,
> gets absorbed by fabrics and what-have-you on the inside, and those items
> then re-radiate the energy in the infrared band.
>
>>
>> Could ultrasonic be the way to go? That's a technology I've yet to play
>> with. It would be used in some sort of RasPi or Arduino configuration.
>
> Sorry -- ultrasonic sensors will reflect off the window itself. You'll
>>
>> Could ultrasonic be the way to go? That's a technology I've yet to play
>> with. It would be used in some sort of RasPi or Arduino configuration.
>
> pretty much have to fit the sensor (for almost anything except image
> processing type detection [this image differs from last image]) outside the
> window and run leads to the R-Pi (or whatever you use to process data).
>
> A PIR sensor /might/ be usable in daylight (they are commonly used for
> auto-activation motion detect security lights). They trigger on an IR
> source moving between segments in the sensor window.
> https://learn.adafruit.com/pir-passive-infrared-proximity-motion-sensor/how-pirs-work
>
> Ultrasonic ("PING)))" -- yes, the ))) is part of the name for some)
> sensors are more distance detection items than motion sensors, and may not
> have a great range. Some old Polaroid cameras used them for focusing -- but
> one has to account for the default focus being "infinity", and
> depth-of-field on those being deep enough that anything around 15+ feet was
> considered "infinity".
>
> https://www.parallax.com/product/ping-ultrasonic-distance-sensor/ (10ft
> range)
> https://learn.adafruit.com/ultrasonic-sonar-distance-sensors (suggests
> 13ft, but recommends 8ft)
>
>
I've got a Raspberry Pi B with a Pi Camera mounted and use the 'motion'
package to record video when motion is detected - through a window.
Video is not very high quality, but you don't need that.
You could use inotify to watch the destination directory and email you
when motion is recorded.
--
Chris Elvidge
England
Chris Elvidge
Dec 24, 2021, 7:16:47 PM12/24/21
to
The only problem is the email delay if your email program doesn't pick
up quickly.
Run this file from /etc/cron.d/inotify
@reboot root /root/bin/inotify
#!/bin/bash
#/root/bin/inotify
sleep 10
inotifywait --quiet --recursive --monitor --event create \
--timefmt "%A %Y-%m-%d %H:%M" --format "%T %e %w %f" \
/home/pi/pics | \
while read change; do
mail -s "Motion video event" os...@osmc-3plus.local <<<"$change"
done
--
Chris Elvidge
England
David Higton
Dec 25, 2021, 3:07:43 PM12/25/21
to
In message <sq31e8$1dhv$1...@gioia.aioe.org>
what you asked for, but might be a good alternative).
Sense the movement of the clapper by optical reflection.
Either paint a white spot on it, or attach reflective tape, somewhere
that a reflective sensor can see. You may find that you need two sensors
so as to avoid the problem where the clapper comes to rest just on the
threshold of one sensor, and, despite hysteresis, you might find it's
changing state frequently and causing false triggers; if so, position a
second sensor such that it can't possibly be on the threshold at the
same time, and look for changes (in either direction) from both sensors.
I'm still looking forward to hearing what you propose as the method to
alert you, whatever the detector is.
David
Chris Schram <chri...@me.com> wrote:
> On 2021-12-23, Axel Berger <Sp...@Berger-Odenthal.De> wrote:
> > Chris Schram wrote:
> > > It would be used in some sort of RasPi or Arduino configuration.
> >
> > That might work, but it's a lot of hassle and complication. The doorbell
> > already is an electric signal. Could you not use that directly?
> >
> Heh! You don't know about my doorbell. It's a brass replica of an antique.
> You twist a knob on the outside of the door which spins a clapper against
> the bell on the inside. No electronics, or even electricity involved.
OK, here's a suggestion for sensing the bell (which, I recognise, isn't
> On 2021-12-23, Axel Berger <Sp...@Berger-Odenthal.De> wrote:
> > Chris Schram wrote:
> > > It would be used in some sort of RasPi or Arduino configuration.
> >
> > That might work, but it's a lot of hassle and complication. The doorbell
> > already is an electric signal. Could you not use that directly?
> >
> Heh! You don't know about my doorbell. It's a brass replica of an antique.
> You twist a knob on the outside of the door which spins a clapper against
> the bell on the inside. No electronics, or even electricity involved.
what you asked for, but might be a good alternative).
Sense the movement of the clapper by optical reflection.
Either paint a white spot on it, or attach reflective tape, somewhere
that a reflective sensor can see. You may find that you need two sensors
so as to avoid the problem where the clapper comes to rest just on the
threshold of one sensor, and, despite hysteresis, you might find it's
changing state frequently and causing false triggers; if so, position a
second sensor such that it can't possibly be on the threshold at the
same time, and look for changes (in either direction) from both sensors.
I'm still looking forward to hearing what you propose as the method to
alert you, whatever the detector is.
David
0 new messages