Comments (9)
pageauc
commented on August 18, 2024
You should not be too close but I think your camera resolution is too big
and therefore is slowing opencv motion detection and causing a timeout.
The faster the objects moves the faster opencv has to track and process
image stream. In your case the camera is very close and the moving
objects will travel very quickly through frame and the processing won't be
fast enough..
I suggest you change resolution to something like 320x240 (default) or
640x480 and see results. Note because these are small resolutions. There
is a *image_bigger =* setting that will increase size of saved image.
There is another one for opencv *WINDOW_BIGGER= *display window resizing .
Also I recommend you NOT try to view opencv image in real time. Set
*gui_window_on = False* This will speed up image processing You can view saved images in
the web browser interface.
I normal turn off the GUI desktop and run speed-cam.py from SSH session
Also I have added a feature in release 10.0 that allows over riding the
automatic calculation of motion crop area If a larger image resolution is
used then a smaller motion crop area can be set that can speed image
processing.
Let me know how you manage.
Claude ...
…On Sat, Oct 19, 2019 at 10:16 AM Matthieu ***@***.***> wrote:
Hi,
I don't get it. Everything looks working correctly but the vehicles looks
not being detected.
The logs says most of the time something like this :
2019-10-19 14:54:34 INFO speed_notify Begin Motion Tracking .....
2019-10-19 15:02:30 INFO speed_camera Reset- event_timer 476.63>0.60 sec
Exceeded 2019-10-19 15:02:30 INFO speed_camera New - 0/10 xy(70,45) Start
New Track 2019-10-19 15:04:19 INFO speed_camera Reset- event_timer
109.05>0.60 sec Exceeded 2019-10-19 15:04:19 INFO speed_camera New - 0/10
xy(74,37) Start New Track 2019-10-19 15:07:27 INFO speed_camera Reset-
event_timer 187.96>0.60 sec Exceeded 2019-10-19 15:07:27 INFO speed_camera
New - 0/10 xy(62,43) Start New Track 2019-10-19 15:07:28 INFO speed_camera
Reset- event_timer 0.70>0.60 sec Exceeded 2019-10-19 15:07:28 INFO
speed_camera New - 0/10 xy(63,43) Start New Track ^C
The only time I saw the system actually measuring something was when a
pedestrian passed in the field 😂
This is the picture I'm capturing. NB : Cars displayed are parked.
[image: image]
<https://user-images.githubusercontent.com/1886799/67146435-5875c580-f28b-11e9-80b1-eff690e966d9.png>
What should I do ? Am I too close ?
Txs
Matthieu
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#44?email_source=notifications&email_token=ABNPKZGAISUAGZ36OVCEP5LQPMJDLA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HS5XBMA>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABNPKZAO6I3EUEYFCDP4LQTQPMJDLANCNFSM4JCP7TBQ>
.
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
from speed-camera.
pagaille
commented on August 18, 2024
First thank you very much for your extensive answer !
I’m running at 320x200 (default settings) but I enabled the GUI to better grasp what was happening. I did that by forwarding X11 through ssh, which is probably harder for the cpu. The screen capture is done on a Retina display that’s why it looks high res.
I’ll try again without the display and by fine tuning the system using the tips you gave me here.
May I suggest to include this tips in the readme ?
Also : I tough, by reading the calibration procedure, that y_upper and y_lower were configurable from the setup.py file. It looks like it is not the case.
—> I’d like to be able to move that detection window : in my case moving the camera is not easy and the detection window doesn’t cover the whole road.
—> I guess the doc should be corrected in the current situation.
Thanks, I’ll report back.
Matthieu G. (en mode mobile)
… Le 19 oct. 2019 à 17:53, Claude Pageau ***@***.***> a écrit :
You should not be too close but I think your camera resolution is too big
and therefore is slowing opencv motion detection and causing a timeout.
The faster the objects moves the faster opencv has to track and process
image stream. In your case the camera is very close and the moving
objects will travel very quickly through frame and the processing won't be
fast enough..
I suggest you change resolution to something like 320x240 (default) or
640x480 and see results. Note because these are small resolutions. There
is *image_bigger =* a setting
that will increase size of saved image. There is another one for
opencv *WINDOW_BIGGER
= *display window resizing .
Also I recommend you NOT try to view opencv image in real time. Set
*gui_window_on =
False* This will speed up image processing You can view saved images in
the web browser interface.
I normal turn off the GUI desktop and run speed-cam.py from SSH session
Also I have added a feature in release 10.0 that allows over riding the
automatic calculation of motion crop area If a larger image resolution is
used then a smaller motion crop area can be set that can speed image
processing.
Let me know how you manage.
Claude ...
On Sat, Oct 19, 2019 at 10:16 AM Matthieu ***@***.***> wrote:
> Hi,
>
> I don't get it. Everything looks working correctly but the vehicles looks
> not being detected.
>
> The logs says most of the time something like this :
>
> 2019-10-19 14:54:34 INFO speed_notify Begin Motion Tracking .....
> 2019-10-19 15:02:30 INFO speed_camera Reset- event_timer 476.63>0.60 sec
> Exceeded 2019-10-19 15:02:30 INFO speed_camera New - 0/10 xy(70,45) Start
> New Track 2019-10-19 15:04:19 INFO speed_camera Reset- event_timer
> 109.05>0.60 sec Exceeded 2019-10-19 15:04:19 INFO speed_camera New - 0/10
> xy(74,37) Start New Track 2019-10-19 15:07:27 INFO speed_camera Reset-
> event_timer 187.96>0.60 sec Exceeded 2019-10-19 15:07:27 INFO speed_camera
> New - 0/10 xy(62,43) Start New Track 2019-10-19 15:07:28 INFO speed_camera
> Reset- event_timer 0.70>0.60 sec Exceeded 2019-10-19 15:07:28 INFO
> speed_camera New - 0/10 xy(63,43) Start New Track ^C
>
> The only time I saw the system actually measuring something was when a
> pedestrian passed in the field 😂
>
> This is the picture I'm capturing. NB : Cars displayed are parked.
>
> [image: image]
> <https://user-images.githubusercontent.com/1886799/67146435-5875c580-f28b-11e9-80b1-eff690e966d9.png>
>
> What should I do ? Am I too close ?
>
> Txs
>
> Matthieu
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <#44?email_source=notifications&email_token=ABNPKZGAISUAGZ36OVCEP5LQPMJDLA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HS5XBMA>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ABNPKZAO6I3EUEYFCDP4LQTQPMJDLANCNFSM4JCP7TBQ>
> .
>
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub <#44?email_source=notifications&email_token=AAOMUT2UV24TXADD2VMNILLQPMUNZA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBXUPEA#issuecomment-544163728>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAOMUT4STFG4KI7TD3DZLILQPMUNZANCNFSM4JCP7TBQ>.
from speed-camera.
pageauc
commented on August 18, 2024
Thanks for your reply. Camera resolution should not be a problem. Note,
You can change the motion tracking area on Release 10.0 of speed-cam.py.
Upgrade on a RPI using menubox.sh UPGRADE menu pick.
The latest config.py file will be called config.py.new. See instructions
at end of UPGRADE to copy config.py.new to config.py. The new config.py
will have a section with rem'd out settings that can over ride the x_left,
x_right, y_upper and y_lower variables so they don't get auto calculated.
This will allow you to customize the motion tracking window.
Please answer some questions
Model of Raspberry PI or computer that you are running speed-cam.py on
Version of operating system Eg Raspbian Buster.
Type of camera
Send me a copy of the config.py that you use for testing below.
Send me a copy of log file (not too long) that you are testing with
verbose is True by default. You can enable sending log to file with
config.py variable *loggingToFile = True*
optionally you can change the file log name. Keep the log fairly short but
log enough to show enough object activity.
Thanks Claude ..
…On Sat, Oct 19, 2019 at 3:24 PM Matthieu ***@***.***> wrote:
First thank you very much for your extensive answer !
I’m running at 320x200 (default settings) but I enabled the GUI to better
grasp what was happening. I did that by forwarding X11 through ssh, which
is probably harder for the cpu. The screen capture is done on a Retina
display that’s why it looks high res.
I’ll try again without the display and by fine tuning the system using the
tips you gave me here.
May I suggest to include this tips in the readme ?
Also : I tough, by reading the calibration procedure, that y_upper and
y_lower were configurable from the setup.py file. It looks like it is not
the case.
—> I’d like to be able to move that detection window : in my case moving
the camera is not easy and the detection window doesn’t cover the whole
road.
—> I guess the doc should be corrected in the current situation.
Thanks, I’ll report back.
Matthieu G. (en mode mobile)
> Le 19 oct. 2019 à 17:53, Claude Pageau ***@***.***> a
écrit :
>
> You should not be too close but I think your camera resolution is too
big
> and therefore is slowing opencv motion detection and causing a timeout.
> The faster the objects moves the faster opencv has to track and process
> image stream. In your case the camera is very close and the moving
> objects will travel very quickly through frame and the processing won't
be
> fast enough..
>
> I suggest you change resolution to something like 320x240 (default) or
> 640x480 and see results. Note because these are small resolutions. There
> is *image_bigger =* a setting
> that will increase size of saved image. There is another one for
> opencv *WINDOW_BIGGER
> = *display window resizing .
>
> Also I recommend you NOT try to view opencv image in real time. Set
> *gui_window_on =
> False* This will speed up image processing You can view saved images in
> the web browser interface.
> I normal turn off the GUI desktop and run speed-cam.py from SSH session
>
> Also I have added a feature in release 10.0 that allows over riding the
> automatic calculation of motion crop area If a larger image resolution is
> used then a smaller motion crop area can be set that can speed image
> processing.
>
> Let me know how you manage.
> Claude ...
>
> On Sat, Oct 19, 2019 at 10:16 AM Matthieu ***@***.***>
wrote:
>
> > Hi,
> >
> > I don't get it. Everything looks working correctly but the vehicles
looks
> > not being detected.
> >
> > The logs says most of the time something like this :
> >
> > 2019-10-19 14:54:34 INFO speed_notify Begin Motion Tracking .....
> > 2019-10-19 15:02:30 INFO speed_camera Reset- event_timer 476.63>0.60
sec
> > Exceeded 2019-10-19 15:02:30 INFO speed_camera New - 0/10 xy(70,45)
Start
> > New Track 2019-10-19 15:04:19 INFO speed_camera Reset- event_timer
> > 109.05>0.60 sec Exceeded 2019-10-19 15:04:19 INFO speed_camera New -
0/10
> > xy(74,37) Start New Track 2019-10-19 15:07:27 INFO speed_camera Reset-
> > event_timer 187.96>0.60 sec Exceeded 2019-10-19 15:07:27 INFO
speed_camera
> > New - 0/10 xy(62,43) Start New Track 2019-10-19 15:07:28 INFO
speed_camera
> > Reset- event_timer 0.70>0.60 sec Exceeded 2019-10-19 15:07:28 INFO
> > speed_camera New - 0/10 xy(63,43) Start New Track ^C
> >
> > The only time I saw the system actually measuring something was when a
> > pedestrian passed in the field 😂
> >
> > This is the picture I'm capturing. NB : Cars displayed are parked.
> >
> > [image: image]
> > <
https://user-images.githubusercontent.com/1886799/67146435-5875c580-f28b-11e9-80b1-eff690e966d9.png
>
> >
> > What should I do ? Am I too close ?
> >
> > Txs
> >
> > Matthieu
> >
> > —
> > You are receiving this because you are subscribed to this thread.
> > Reply to this email directly, view it on GitHub
> > <
#44?email_source=notifications&email_token=ABNPKZGAISUAGZ36OVCEP5LQPMJDLA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HS5XBMA
>,
> > or unsubscribe
> > <
https://github.com/notifications/unsubscribe-auth/ABNPKZAO6I3EUEYFCDP4LQTQPMJDLANCNFSM4JCP7TBQ
>
> > .
> >
>
>
> --
> See my YouTube Channel at http://www.youtube.com/user/pageaucp
> —
> You are receiving this because you authored the thread.
> Reply to this email directly, view it on GitHub <
#44?email_source=notifications&email_token=AAOMUT2UV24TXADD2VMNILLQPMUNZA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBXUPEA#issuecomment-544163728>,
or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAOMUT4STFG4KI7TD3DZLILQPMUNZANCNFSM4JCP7TBQ
>.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#44?email_source=notifications&email_token=ABNPKZAPEGNPKSGRVGVC2DTQPNNFFA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBX23XA#issuecomment-544189916>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABNPKZFMP3MJG6SIXEXGRXTQPNNFFANCNFSM4JCP7TBQ>
.
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
from speed-camera.
pageauc
commented on August 18, 2024
Mathieu
Was thinking image appears to be tool big and wrong aspect ratio to be
320x240.
Check config.py and set
*pluginEnable = False*
Not sure if you are using webcam or picam but make sure it is set to
320x240. You should bet a squarer image 4/3 aspect ratio image.
*image_bigger
= 3.0* will resize the saved image by a factor of 3 but this can be changed
if you want smaller or larger saved image.
See the new 10.0 config.py and you can override the default motion tracking
area per comments for setting* x_left, x_right, y_upper, y_lower*
Restart speed-cam.py
Test and let me know results per previous email
Thanks
Claude ...
…On Sat, Oct 19, 2019 at 4:17 PM Claude Pageau ***@***.***> wrote:
Thanks for your reply. Camera resolution should not be a problem. Note,
You can change the motion tracking area on Release 10.0 of speed-cam.py.
Upgrade on a RPI using menubox.sh UPGRADE menu pick.
The latest config.py file will be called config.py.new. See instructions
at end of UPGRADE to copy config.py.new to config.py. The new config.py
will have a section with rem'd out settings that can over ride the x_left,
x_right, y_upper and y_lower variables so they don't get auto calculated.
This will allow you to customize the motion tracking window.
Please answer some questions
Model of Raspberry PI or computer that you are running speed-cam.py on
Version of operating system Eg Raspbian Buster.
Type of camera
Send me a copy of the config.py that you use for testing below.
Send me a copy of log file (not too long) that you are testing with
verbose is True by default. You can enable sending log to file with
config.py variable *loggingToFile = True*
optionally you can change the file log name. Keep the log fairly short
but log enough to show enough object activity.
Thanks Claude ..
On Sat, Oct 19, 2019 at 3:24 PM Matthieu ***@***.***> wrote:
> First thank you very much for your extensive answer !
>
> I’m running at 320x200 (default settings) but I enabled the GUI to better
> grasp what was happening. I did that by forwarding X11 through ssh, which
> is probably harder for the cpu. The screen capture is done on a Retina
> display that’s why it looks high res.
>
> I’ll try again without the display and by fine tuning the system using
> the tips you gave me here.
>
> May I suggest to include this tips in the readme ?
>
> Also : I tough, by reading the calibration procedure, that y_upper and
> y_lower were configurable from the setup.py file. It looks like it is not
> the case.
>
> —> I’d like to be able to move that detection window : in my case moving
> the camera is not easy and the detection window doesn’t cover the whole
> road.
> —> I guess the doc should be corrected in the current situation.
>
> Thanks, I’ll report back.
>
> Matthieu G. (en mode mobile)
>
> > Le 19 oct. 2019 à 17:53, Claude Pageau ***@***.***> a
> écrit :
> >
> > You should not be too close but I think your camera resolution is too
> big
> > and therefore is slowing opencv motion detection and causing a timeout.
> > The faster the objects moves the faster opencv has to track and process
> > image stream. In your case the camera is very close and the moving
> > objects will travel very quickly through frame and the processing won't
> be
> > fast enough..
> >
> > I suggest you change resolution to something like 320x240 (default) or
> > 640x480 and see results. Note because these are small resolutions. There
> > is *image_bigger =* a setting
> > that will increase size of saved image. There is another one for
> > opencv *WINDOW_BIGGER
> > = *display window resizing .
> >
> > Also I recommend you NOT try to view opencv image in real time. Set
> > *gui_window_on =
> > False* This will speed up image processing You can view saved images in
> > the web browser interface.
> > I normal turn off the GUI desktop and run speed-cam.py from SSH session
> >
> > Also I have added a feature in release 10.0 that allows over riding the
> > automatic calculation of motion crop area If a larger image resolution
> is
> > used then a smaller motion crop area can be set that can speed image
> > processing.
> >
> > Let me know how you manage.
> > Claude ...
> >
> > On Sat, Oct 19, 2019 at 10:16 AM Matthieu ***@***.***>
> wrote:
> >
> > > Hi,
> > >
> > > I don't get it. Everything looks working correctly but the vehicles
> looks
> > > not being detected.
> > >
> > > The logs says most of the time something like this :
> > >
> > > 2019-10-19 14:54:34 INFO speed_notify Begin Motion Tracking .....
> > > 2019-10-19 15:02:30 INFO speed_camera Reset- event_timer 476.63>0.60
> sec
> > > Exceeded 2019-10-19 15:02:30 INFO speed_camera New - 0/10 xy(70,45)
> Start
> > > New Track 2019-10-19 15:04:19 INFO speed_camera Reset- event_timer
> > > 109.05>0.60 sec Exceeded 2019-10-19 15:04:19 INFO speed_camera New -
> 0/10
> > > xy(74,37) Start New Track 2019-10-19 15:07:27 INFO speed_camera Reset-
> > > event_timer 187.96>0.60 sec Exceeded 2019-10-19 15:07:27 INFO
> speed_camera
> > > New - 0/10 xy(62,43) Start New Track 2019-10-19 15:07:28 INFO
> speed_camera
> > > Reset- event_timer 0.70>0.60 sec Exceeded 2019-10-19 15:07:28 INFO
> > > speed_camera New - 0/10 xy(63,43) Start New Track ^C
> > >
> > > The only time I saw the system actually measuring something was when a
> > > pedestrian passed in the field 😂
> > >
> > > This is the picture I'm capturing. NB : Cars displayed are parked.
> > >
> > > [image: image]
> > > <
> https://user-images.githubusercontent.com/1886799/67146435-5875c580-f28b-11e9-80b1-eff690e966d9.png
> >
> > >
> > > What should I do ? Am I too close ?
> > >
> > > Txs
> > >
> > > Matthieu
> > >
> > > —
> > > You are receiving this because you are subscribed to this thread.
> > > Reply to this email directly, view it on GitHub
> > > <
> #44?email_source=notifications&email_token=ABNPKZGAISUAGZ36OVCEP5LQPMJDLA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HS5XBMA
> >,
> > > or unsubscribe
> > > <
> https://github.com/notifications/unsubscribe-auth/ABNPKZAO6I3EUEYFCDP4LQTQPMJDLANCNFSM4JCP7TBQ
> >
> > > .
> > >
> >
> >
> > --
> > See my YouTube Channel at http://www.youtube.com/user/pageaucp
> > —
> > You are receiving this because you authored the thread.
> > Reply to this email directly, view it on GitHub <
> #44?email_source=notifications&email_token=AAOMUT2UV24TXADD2VMNILLQPMUNZA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBXUPEA#issuecomment-544163728>,
> or unsubscribe <
> https://github.com/notifications/unsubscribe-auth/AAOMUT4STFG4KI7TD3DZLILQPMUNZANCNFSM4JCP7TBQ
> >.
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#44?email_source=notifications&email_token=ABNPKZAPEGNPKSGRVGVC2DTQPNNFFA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBX23XA#issuecomment-544189916>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ABNPKZFMP3MJG6SIXEXGRXTQPNNFFANCNFSM4JCP7TBQ>
> .
>
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
from speed-camera.
pagaille
commented on August 18, 2024
Hi Claude, txs again for your help.
Model of Raspberry PI or computer that you are running speed-cam.py on
Pi3B+
Version of operating system Eg Raspbian Buster.
Yep, Buster, updated.
Type of camera
Good question. That's a subminiature pi camera 'spy camera' that I had on the shelf. It is based on the OV5647. It's field of view is "54 x 41 degrees"
I bought it here : https://shop.mchobby.be/fr/pi-camera/1117-camera-raspberry-pi-espion-spy-5-megapixel-3232100011175.html
And it's probably that same model : https://www.arducam.com/product/1-4-5-noir-mpspy-camera-module-sensor-flex-cable-raspberry-pi/
Regarding the resolution and aspect ratio, I'm not sure. It could be linked to the X11 ssh forward and my mac display.
I tested again with version 10 and disabled the UI. I basically get the same results : slow objects are correctly detected and tracked, I get the 5 calibration pictures of a pedestrian passing by.
But I only get one picture or two of cars passing in front of my house. It looks like the fps isn't good enough. What could I do ?
from speed-camera.
pagaille
commented on August 18, 2024
This is typically the result I get :
- Contour is wrong
- Only one picture recorded in calibration mode
- speed is wrong (and somewhat random)
NB The black Caddy is mine and is parked :-)
----------------------------------------------------------------------
2019-10-23 10:20:18 INFO speed_camera Reset- event_timer 274.78>0.30 sec Exceeded
2019-10-23 10:20:18 INFO speed_camera New - 0/5 xy(73,122) Start New Track
2019-10-23 10:20:18 INFO speed_camera Out - 0/5 xy(73,121) Min D=0<=1px C=2 27x39=518 sqpx R2L
2019-10-23 10:20:18 INFO speed_camera New - 0/5 xy(131,112) Start New Track
2019-10-23 10:20:18 INFO speed_camera Out - 0/5 xy(194,91) Max D=63>=20px C=2 17x11=111 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Reset- event_timer 96.65>0.30 sec Exceeded
2019-10-23 10:21:55 INFO speed_camera New - 0/5 xy(30,16) Start New Track
2019-10-23 10:21:55 INFO speed_camera Add - 1/5 xy(46,17) 46.98 kph D=16/20 C=2 158x91=6496 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Add - 2/5 xy(64,18) 55.14 kph D=18/20 C=1 156x82=6133 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Add - 3/5 xy(78,18) 21.20 kph D=14/20 C=6 164x106=7707 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Add - 4/5 xy(97,19) 56.09 kph D=19/20 C=1 152x82=6061 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Out - 4/5 xy(147,79) Max D=50>=20px C=3 21x14=167 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Out - 4/5 xy(205,86) Max D=58>=20px C=2 15x17=142 sqpx L2R
2019-10-23 10:21:55 INFO speed_camera Add - 5/5 xy(219,87) 3.84 kph D=14/20 C=2 18x18=185 sqpx L2R
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1021559.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 36.6 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1021559.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
Config.py :
# ---------------- User Configuration Settings for speed-cam.py ---------------------------------
# Ver 10.0 speed-cam.py Variable Configuration Settings
#######################################
# speed-cam.py Variable Settings
# Default is 320x240 image stream size
# Note motion track crop area is now
# auto configured so you may not need to use a plugin
# to change resolution.
#######################################
# Calibration Settings
# --------------------
calibrate = True # Create a calibration image file with calibration hash markers 10 px per mark
cal_obj_px_L2R = 150 # L2R Moving Objects, Length of a calibration object in pixels
cal_obj_mm_L2R = 4408 # L2R Moving Objects, Length of the calibration object in millimetres
cal_obj_px_R2L = 150 # R2L Moving Objects, Length of a calibration object in pixels
cal_obj_mm_R2L = 4408 # R2L Moving Objects, Length of the calibration object in millimetres
# Note if tested speed is too low increase appropriate cal_obj_mm value and redo speed test for desired direction.
# IMPORTANT - If plugins Enabled Edit Settings in specified plugin file located in plugins folder.
# Plugins overlay the config.py variable settings
# -----------------------------------------------
pluginEnable = False
pluginName = "picam240" # Specify filename in plugins subfolder without .py extension per below
# picam240, webcam240 (Recommended for RPI2 or greater)
# picam480, webcam480, picam720, webcam720 (can use RPI3 but Test)
# picam1080 (Experimental Not Recommended)
# secpicam480, secwebcam480 (Experimental no CSV entries)
# Display opencv windows on gui desktop
# gui_window_on suppresses All Windows if False
# ----------------------------------------------
gui_window_on = False # True= Turn On All desktop GUI openCV windows. False=Don't Show (req'd for SSH) .
show_thresh_on = False # Display desktop GUI openCV cropped threshold window. True=Show, False=Don't Show
show_crop_on = False # Same as show_thresh_on but in color. True=Show, False=Don't Show (Default)
# Display and Log settings
# ------------------------
verbose = True # True= Display basic status information on console False= Off
display_fps = False # True= Show average frame count every 1000 loops False= Off
log_data_to_CSV = False # True= Save log data as CSV comma separated values False= Off
loggingToFile = False # True= Send logging to file False= No Logging to File
logFilePath = 'speed-cam.log' # Location of log file when loggingToFile=True
# Motion Event Settings
# ---------------------
SPEED_MPH = False # Set Speed Units kph=False mph=True
track_counter = 5 # one cpu core=5 four cores=10 Number of Consecutive Motion Events to trigger speed photo
# Testing with RPI2 B one core using opencv ver 3.4.2
MIN_AREA = 100 # Default= 100 Exclude all contours less than or equal to this sq-px Area
show_out_range = True # Default= True Show Out of Range Events per x_diff settings below False= Off
x_diff_max = 20 # Default= 20 Exclude if max px away >= last motion event x position
x_diff_min = 1 # Default= 1 Exclude if min px away <= last event x position
x_buf_adjust = 10 # Default= 10 Divides motion Rect x for L&R Buffer Space to Ensure contours are in
track_timeout = 0.0 # Default= 0.0 Optional seconds to wait after track End (Avoids dual tracking)
event_timeout = 0.3 # Default= 0.3 seconds to wait for next motion event before starting new track
max_speed_over = 0 # Exclude track if Speed less than or equal to value specified 0=All
# Can be useful to exclude pedestrians and/or bikes, Etc or track only fast objects
# Note: To see motion tracking crop area on images, Set variable image_show_motion_area = True
# Allow user to customize the motion crop area (width) x values
# If variables not found then values will be set automatically base on image size.
x_left = 20 # uncomment and change values to override auto calculate
x_right = 300 # uncomment and change values to override auto calculate
# Allow user to customize the motion crop area (height) y values
# If variables not found then values will be set automatically base on image size.
y_upper = 80 # uncomment and change values to override auto calculate
y_lower = 240 # uncomment and change values to override auto calculate
# Camera Settings
# ---------------
WEBCAM = False # Default = False False=PiCamera True=USB WebCamera
# Web Camera Settings
WEBCAM_SRC = 0 # Default= 0 USB opencv connection number
# or ip cam rtsp string eg "rtsp://192.168.1.101/RtspTranslator.12/camera"
# See ip cam docs. WEBCAM_WIDTH & HEIGHT MUST match rtsp stream resolution
WEBCAM_WIDTH = 320 # Default= 320 USB Webcam Image width
WEBCAM_HEIGHT = 240 # Default= 240 USB Webcam Image height
WEBCAM_HFLIP = False # Default= False USB Webcam flip image horizontally
WEBCAM_VFLIP = False # Default= False USB Webcam flip image vertically
# IMPORTANT Webcam Streaming Performance Hit if Stream Flipped.
# Pi Camera Settings
# ------------------
CAMERA_WIDTH = 320 # Image stream width for opencv motion scanning Default=320
CAMERA_HEIGHT = 240 # Image stream height for opencv motion scanning Default=240
CAMERA_FRAMERATE = 20 # Default= 20 Frame rate for video stream V2 picam can be higher
CAMERA_ROTATION = 0 # Rotate camera image valid values are 0, 90, 180, 270
CAMERA_VFLIP = True # Flip the camera image vertically if required
CAMERA_HFLIP = True # Flip the camera image horizontally if required
# Camera Image Settings
# ---------------------
image_path = "media/images" # folder name to store images
image_prefix = "speed-" # image name prefix
image_format = ".jpg" # Default = ".jpg" image Formats .jpeg .png .gif .bmp
image_show_motion_area = True # True= Display motion detection rectangle area on saved images
image_filename_speed = False # True= Prefix filename with speed value
image_text_on = True # True= Show Text on speed images False= No Text on images
image_text_bottom = True # True= Show image text at bottom otherwise at top
image_font_size = 12 # Default= 12 Font text height in px for text on images
image_bigger = 3.0 # Default= 3.0 min=0.1 Resize saved speed image by specified multiplier value
image_max_files = 0 # 0=off or specify MaxFiles to maintain then oldest are deleted Default=0 (off)
# Optional Manage SubDir Creation by time, number of files or both (not recommended)
# ----------------------------------------------------------------
imageSubDirMaxFiles = 1000 # 0=off or specify MaxFiles - Creates New dated sub-folder if MaxFiles exceeded
imageSubDirMaxHours = 0 # 0=off or specify MaxHours - Creates New dated sub-folder if MaxHours exceeded
# Optional Save Most Recent files in recent folder
# ------------------------------------------------
imageRecentMax = 100 # 0=off Maintain specified number of most recent files in motionRecentDir
imageRecentDir = "media/recent" # Default= "media/recent" save recent files directory path
# Optional Manage Free Disk Space Settings
# ----------------------------------------
spaceTimerHrs = 0 # Default= 0 0=off or specify hours frequency to perform free disk space check
spaceFreeMB = 500 # Default= 500 Target Free space in MB Required.
spaceMediaDir = 'media/images' # Default= 'media/images' Starting point for directory walk
spaceFileExt = 'jpg' # Default= 'jpg' File extension to Delete Oldest Files
# OpenCV Motion Settings
# ----------------------
SHOW_CIRCLE = False # True=circle in center of motion, False=rectangle
CIRCLE_SIZE = 5 # Default= 5 Diameter circle in px if SHOW_CIRCLE = True
LINE_THICKNESS = 1 # Default= 1 Size of lines for circle or Rectangle
FONT_SCALE = 0.5 # Default= 0.5 OpenCV window text font size scaling factor Default=.5 (lower is smaller)
WINDOW_BIGGER = 1.0 # Default= 1.0 min=0.1 Resize multiplier for opencv window if gui_window_on=True
BLUR_SIZE = 10 # Default= 10 OpenCV setting for Gaussian difference image blur
THRESHOLD_SENSITIVITY = 20 # Default= 20 OpenCV setting for difference image threshold
# Sqlite3 Settings
# ----------------
DB_DIR = "data"
DB_NAME = "speed_cam.db"
DB_TABLE = "speed"
#======================================
# webserver.py Settings
#======================================
# Web Server settings
# -------------------
web_server_port = 8080 # Default= 8080 Web server access port eg http://192.168.1.100:8080
web_server_root = "media" # Default= "media" webserver root path to webserver image/video sub-folders
web_page_title = "SPEED-CAMERA Media" # web page title that browser show (not displayed on web page)
web_page_refresh_on = True # False=Off (never) Refresh True=On (per seconds below)
web_page_refresh_sec = "900" # Default= "900" seconds to wait for web page refresh seconds (15 minutes)
web_page_blank = False # True Starts left image with a blank page until a right menu item is selected
# False displays second list[1] item since first may be in progress
# Left iFrame Image Settings
# --------------------------
web_image_height = "768" # Default= "768" px height of images to display in iframe
web_iframe_width_usage = "70%" # Left Pane - Sets % of total screen width allowed for iframe. Rest for right list
web_iframe_width = "100%" # Desired frame width to display images. can be eg percent "80%" or px "1280"
web_iframe_height = "100%" # Desired frame height to display images. Scroll bars if image larger (percent or px)
# Right Side Files List
# ---------------------
web_max_list_entries = 0 # 0 = All or Specify Max right side file entries to show (must be > 1)
web_list_height = web_image_height # Right List - side menu height in px (link selection)
web_list_by_datetime = True # True=datetime False=filename
web_list_sort_descending = True # reverse sort order (filename or datetime per web_list_by_datetime setting
# ---------------------------------------------- End of User Variables -----------------------------------------------------
from speed-camera.
pagaille
commented on August 18, 2024
I tried one more try with my car. Neighbours start wondering what I'm doing :-)
I cruised at 24kph.
I got four pictures :
And here is the corresponding log :
----------------------------------------------------------------------
2019-10-23 10:38:22 INFO speed_camera New - 0/5 xy(164,61) Start New Track
2019-10-23 10:38:22 INFO speed_camera Add - 1/5 xy(179,61) 35.97 kph D=15/20 C=6 39x85=1717 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 2/5 xy(186,60) 22.00 kph D=7/20 C=6 40x85=1802 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 3/5 xy(193,59) 10.70 kph D=7/20 C=6 41x85=1834 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 4/5 xy(200,58) 40.50 kph D=7/20 C=5 42x86=1915 sqpx L2R
2019-10-23 10:38:54 INFO speed_camera Reset- event_timer 31.95>0.30 sec Exceeded
2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(138,13) Start New Track
2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(130,13) 23.66 kph D=8/20 C=2 116x72=3544 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(120,12) 29.97 kph D=10/20 C=3 101x73=3367 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(111,12) 13.38 kph D=9/20 C=5 95x72=3059 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(102,12) 25.60 kph D=9/20 C=3 100x72=3333 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 5/5 xy(93,11) 26.68 kph D=9/20 C=3 96x73=3312 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038547.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 23.9 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038547.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:54 INFO speed_camera Saved media/images/calib-20191023-1038547.jpg
2019-10-23 10:38:54 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO speed_camera End - R2L Ave Speed 23.9 kph Tracked 45 px in 0.212 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(84,11) Start New Track
2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(75,10) 49.80 kph D=9/20 C=3 98x73=3334 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(66,11) 27.48 kph D=9/20 C=3 101x74=3365 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(56,10) 15.41 kph D=10/20 C=3 105x75=3203 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(47,11) 26.80 kph D=9/20 C=3 105x72=2919 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(37,10) 29.74 kph D=10/20 C=3 111x75=2915 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038550.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 29.8 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038550.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038550.jpg
2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 29.9 kph Tracked 47 px in 0.191 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(133,11) Start New Track
2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(124,11) 49.99 kph D=9/20 C=3 46x77=1589 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(114,11) 28.73 kph D=10/20 C=4 47x76=1610 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(104,10) 15.32 kph D=10/20 C=4 48x76=1637 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(95,9) 26.23 kph D=9/20 C=4 46x75=1755 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(85,9) 31.31 kph D=10/20 C=5 48x77=1808 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038552.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 30.3 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038552.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038552.jpg
2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 30.3 kph Tracked 48 px in 0.192 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(75,9) Start New Track
2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(65,10) 48.24 kph D=10/20 C=3 49x75=1678 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(56,13) 28.11 kph D=9/20 C=4 48x70=1415 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(46,13) 15.27 kph D=10/20 C=3 49x70=1521 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(55,17) 25.90 kph D=9/20 C=6 30x46=787 sqpx L2R
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(46,16) 26.50 kph D=9/20 C=4 33x47=907 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038555.jpg image for camera calibration
No
from speed-camera.
pageauc
commented on August 18, 2024
Output looks OK. You can try increasing camera FPS for camera to maybe 30
or event slightly more. Opencv needs processing time depending on the
model of RPI so there is a maximum where there is no advantage to
increasing any more. Due to the camera being so close you are getting
variations in speed due to contours being picked up at different locations
due to lighting, reflections etc. If object is farther away the object
appears smaller and these variations are smaller. You also get variations
since camera pixels at edges of image are farther away than ones in middle
of view. This also can cause variations in calculations.
Increase camera framerate .to 30 or 35 (config.py)
You could also try increasing track_counter from 5 to slightly higher
(test) This is also in config.py under motion event settings
Also still not sure of camera resolution. What does it say on settings
page when you run from console with verbose on. I am thinking camera is
not same as RPI camera and not compatible with RPI firmware and not setting
resolution per pi-camera python library.
Claude ...
…On Wed, Oct 23, 2019 at 5:44 AM Matthieu ***@***.***> wrote:
I tried one more try with my car. Neighbours start wondering what I'm
doing :-)
I cruised at 24kph.
I got three pictures :
[image: calib-20191023-1038547]
<https://user-images.githubusercontent.com/1886799/67380252-54e38680-f58a-11e9-8d60-35754719359f.jpg>
[image: calib-20191023-1038550]
<https://user-images.githubusercontent.com/1886799/67380253-54e38680-f58a-11e9-95e8-35279c376189.jpg>
[image: calib-20191023-1038552]
<https://user-images.githubusercontent.com/1886799/67380254-54e38680-f58a-11e9-91d4-f2cd7f40ca7c.jpg>
[image: calib-20191023-1038555]
<https://user-images.githubusercontent.com/1886799/67380255-557c1d00-f58a-11e9-90df-5abfbab4cb61.jpg>
And here is the corresponding log :
----------------------------------------------------------------------
2019-10-23 10:38:22 INFO speed_camera New - 0/5 xy(164,61) Start New Track
2019-10-23 10:38:22 INFO speed_camera Add - 1/5 xy(179,61) 35.97 kph D=15/20 C=6 39x85=1717 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 2/5 xy(186,60) 22.00 kph D=7/20 C=6 40x85=1802 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 3/5 xy(193,59) 10.70 kph D=7/20 C=6 41x85=1834 sqpx L2R
2019-10-23 10:38:22 INFO speed_camera Add - 4/5 xy(200,58) 40.50 kph D=7/20 C=5 42x86=1915 sqpx L2R
2019-10-23 10:38:54 INFO speed_camera Reset- event_timer 31.95>0.30 sec Exceeded
2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(138,13) Start New Track
2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(130,13) 23.66 kph D=8/20 C=2 116x72=3544 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(120,12) 29.97 kph D=10/20 C=3 101x73=3367 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(111,12) 13.38 kph D=9/20 C=5 95x72=3059 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(102,12) 25.60 kph D=9/20 C=3 100x72=3333 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 5/5 xy(93,11) 26.68 kph D=9/20 C=3 96x73=3312 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038547.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 23.9 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038547.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:54 INFO speed_camera Saved media/images/calib-20191023-1038547.jpg
2019-10-23 10:38:54 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:54 INFO speed_camera End - R2L Ave Speed 23.9 kph Tracked 45 px in 0.212 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(84,11) Start New Track
2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(75,10) 49.80 kph D=9/20 C=3 98x73=3334 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(66,11) 27.48 kph D=9/20 C=3 101x74=3365 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(56,10) 15.41 kph D=10/20 C=3 105x75=3203 sqpx R2L
2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(47,11) 26.80 kph D=9/20 C=3 105x72=2919 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(37,10) 29.74 kph D=10/20 C=3 111x75=2915 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038550.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 29.8 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038550.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038550.jpg
2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 29.9 kph Tracked 47 px in 0.191 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(133,11) Start New Track
2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(124,11) 49.99 kph D=9/20 C=3 46x77=1589 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(114,11) 28.73 kph D=10/20 C=4 47x76=1610 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(104,10) 15.32 kph D=10/20 C=4 48x76=1637 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(95,9) 26.23 kph D=9/20 C=4 46x75=1755 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(85,9) 31.31 kph D=10/20 C=5 48x77=1808 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038552.jpg image for camera calibration
Note: If there is only one lane then L2R and R2L settings will be the same
1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
If Recorded Speed 30.3 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
Repeat Calibration with same object moving R2L and update config.py R2L variables
cal_obj_mm_R2L and cal_obj_px_R2L accordingly
4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
6 - When Calibration is Finished, Set config.py Variable calibrate = False
Then Restart speed-cam.py and monitor activity.
WARNING: It is Advised to Use 320x240 Stream for Best Performance.
Higher Resolutions Need More OpenCV Processing
and May Reduce Data Accuracy and Reliability.
Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038552.jpg
View Calibration Image in Web Browser (Ensure webserver.py is started)
---------------------- Press cntl-c to Quit Calibration Mode -----------------------
2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038552.jpg
2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 30.3 kph Tracked 48 px in 0.192 sec Calib 150px 4408mm
----------------------------------------------------------------------
2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(75,9) Start New Track
2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(65,10) 48.24 kph D=10/20 C=3 49x75=1678 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(56,13) 28.11 kph D=9/20 C=4 48x70=1415 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(46,13) 15.27 kph D=10/20 C=3 49x70=1521 sqpx R2L
2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(55,17) 25.90 kph D=9/20 C=6 30x46=787 sqpx L2R
2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(46,16) 26.50 kph D=9/20 C=4 33x47=907 sqpx R2L
----------------------------- Create Calibration Image -----------------------------
Instructions for using media/images/calib-20191023-1038555.jpg image for camera calibration
No
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#44?email_source=notifications&email_token=ABNPKZDSBWTJ6DZ3QIELIJTQQAMHBA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECAY5XI#issuecomment-545361629>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABNPKZDDHVEPYGSFA7HASWDQQAMHBANCNFSM4JCP7TBQ>
.
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
from speed-camera.
pageauc
commented on August 18, 2024
Ok camera is OK and is at 320x240 per hash marks on cal image. Claude ..
…On Wed, Oct 23, 2019 at 7:36 AM Claude Pageau ***@***.***> wrote:
Output looks OK. You can try increasing camera FPS for camera to maybe 30
or event slightly more. Opencv needs processing time depending on the
model of RPI so there is a maximum where there is no advantage to
increasing any more. Due to the camera being so close you are getting
variations in speed due to contours being picked up at different locations
due to lighting, reflections etc. If object is farther away the object
appears smaller and these variations are smaller. You also get variations
since camera pixels at edges of image are farther away than ones in middle
of view. This also can cause variations in calculations.
Increase camera framerate .to 30 or 35 (config.py)
You could also try increasing track_counter from 5 to slightly higher
(test) This is also in config.py under motion event settings
Also still not sure of camera resolution. What does it say on settings
page when you run from console with verbose on. I am thinking camera is
not same as RPI camera and not compatible with RPI firmware and not setting
resolution per pi-camera python library.
Claude ...
On Wed, Oct 23, 2019 at 5:44 AM Matthieu ***@***.***> wrote:
> I tried one more try with my car. Neighbours start wondering what I'm
> doing :-)
>
> I cruised at 24kph.
>
> I got three pictures :
>
> [image: calib-20191023-1038547]
> <https://user-images.githubusercontent.com/1886799/67380252-54e38680-f58a-11e9-8d60-35754719359f.jpg>
> [image: calib-20191023-1038550]
> <https://user-images.githubusercontent.com/1886799/67380253-54e38680-f58a-11e9-95e8-35279c376189.jpg>
> [image: calib-20191023-1038552]
> <https://user-images.githubusercontent.com/1886799/67380254-54e38680-f58a-11e9-91d4-f2cd7f40ca7c.jpg>
> [image: calib-20191023-1038555]
> <https://user-images.githubusercontent.com/1886799/67380255-557c1d00-f58a-11e9-90df-5abfbab4cb61.jpg>
>
> And here is the corresponding log :
>
> ----------------------------------------------------------------------
> 2019-10-23 10:38:22 INFO speed_camera New - 0/5 xy(164,61) Start New Track
> 2019-10-23 10:38:22 INFO speed_camera Add - 1/5 xy(179,61) 35.97 kph D=15/20 C=6 39x85=1717 sqpx L2R
> 2019-10-23 10:38:22 INFO speed_camera Add - 2/5 xy(186,60) 22.00 kph D=7/20 C=6 40x85=1802 sqpx L2R
> 2019-10-23 10:38:22 INFO speed_camera Add - 3/5 xy(193,59) 10.70 kph D=7/20 C=6 41x85=1834 sqpx L2R
> 2019-10-23 10:38:22 INFO speed_camera Add - 4/5 xy(200,58) 40.50 kph D=7/20 C=5 42x86=1915 sqpx L2R
> 2019-10-23 10:38:54 INFO speed_camera Reset- event_timer 31.95>0.30 sec Exceeded
> 2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(138,13) Start New Track
> 2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(130,13) 23.66 kph D=8/20 C=2 116x72=3544 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(120,12) 29.97 kph D=10/20 C=3 101x73=3367 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(111,12) 13.38 kph D=9/20 C=5 95x72=3059 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(102,12) 25.60 kph D=9/20 C=3 100x72=3333 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 5/5 xy(93,11) 26.68 kph D=9/20 C=3 96x73=3312 sqpx R2L
> ----------------------------- Create Calibration Image -----------------------------
>
> Instructions for using media/images/calib-20191023-1038547.jpg image for camera calibration
>
> Note: If there is only one lane then L2R and R2L settings will be the same
> 1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
> 2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
> 3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
> If Recorded Speed 23.9 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
> Repeat Calibration with same object moving R2L and update config.py R2L variables
> cal_obj_mm_R2L and cal_obj_px_R2L accordingly
> 4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
> 5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
> 6 - When Calibration is Finished, Set config.py Variable calibrate = False
> Then Restart speed-cam.py and monitor activity.
>
> WARNING: It is Advised to Use 320x240 Stream for Best Performance.
> Higher Resolutions Need More OpenCV Processing
> and May Reduce Data Accuracy and Reliability.
>
> Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038547.jpg
> View Calibration Image in Web Browser (Ensure webserver.py is started)
>
> ---------------------- Press cntl-c to Quit Calibration Mode -----------------------
> 2019-10-23 10:38:54 INFO speed_camera Saved media/images/calib-20191023-1038547.jpg
> 2019-10-23 10:38:54 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:54 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:54 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:54 INFO speed_camera End - R2L Ave Speed 23.9 kph Tracked 45 px in 0.212 sec Calib 150px 4408mm
> ----------------------------------------------------------------------
> 2019-10-23 10:38:54 INFO speed_camera New - 0/5 xy(84,11) Start New Track
> 2019-10-23 10:38:54 INFO speed_camera Add - 1/5 xy(75,10) 49.80 kph D=9/20 C=3 98x73=3334 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 2/5 xy(66,11) 27.48 kph D=9/20 C=3 101x74=3365 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 3/5 xy(56,10) 15.41 kph D=10/20 C=3 105x75=3203 sqpx R2L
> 2019-10-23 10:38:54 INFO speed_camera Add - 4/5 xy(47,11) 26.80 kph D=9/20 C=3 105x72=2919 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(37,10) 29.74 kph D=10/20 C=3 111x75=2915 sqpx R2L
> ----------------------------- Create Calibration Image -----------------------------
>
> Instructions for using media/images/calib-20191023-1038550.jpg image for camera calibration
>
> Note: If there is only one lane then L2R and R2L settings will be the same
> 1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
> 2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
> 3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
> If Recorded Speed 29.8 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
> Repeat Calibration with same object moving R2L and update config.py R2L variables
> cal_obj_mm_R2L and cal_obj_px_R2L accordingly
> 4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
> 5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
> 6 - When Calibration is Finished, Set config.py Variable calibrate = False
> Then Restart speed-cam.py and monitor activity.
>
> WARNING: It is Advised to Use 320x240 Stream for Best Performance.
> Higher Resolutions Need More OpenCV Processing
> and May Reduce Data Accuracy and Reliability.
>
> Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038550.jpg
> View Calibration Image in Web Browser (Ensure webserver.py is started)
>
> ---------------------- Press cntl-c to Quit Calibration Mode -----------------------
> 2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038550.jpg
> 2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 29.9 kph Tracked 47 px in 0.191 sec Calib 150px 4408mm
> ----------------------------------------------------------------------
> 2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(133,11) Start New Track
> 2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(124,11) 49.99 kph D=9/20 C=3 46x77=1589 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(114,11) 28.73 kph D=10/20 C=4 47x76=1610 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(104,10) 15.32 kph D=10/20 C=4 48x76=1637 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(95,9) 26.23 kph D=9/20 C=4 46x75=1755 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(85,9) 31.31 kph D=10/20 C=5 48x77=1808 sqpx R2L
> ----------------------------- Create Calibration Image -----------------------------
>
> Instructions for using media/images/calib-20191023-1038552.jpg image for camera calibration
>
> Note: If there is only one lane then L2R and R2L settings will be the same
> 1 - Use L2R and R2L with Same Size Reference Object, Eg. same vehicle for both directions.
> 2 - For objects moving L2R Record cal_obj_px_L2R Value Using Red y_upper Hash Marks at every 10 px Current Setting is 150 px
> 3 - Record cal_obj_mm_L2R of object. This is Actual length in mm of object above Current Setting is 4408 mm
> If Recorded Speed 30.3 kph is Too Low, Increasing cal_obj_mm_L2R to Adjust or Visa-Versa
> Repeat Calibration with same object moving R2L and update config.py R2L variables
> cal_obj_mm_R2L and cal_obj_px_R2L accordingly
> 4 - Edit /home/pi/speed-camera/config.py File and Change Values for the Above Variables.
> 5 - Do a Speed Test to Confirm/Tune Settings. You May Need to Repeat.
> 6 - When Calibration is Finished, Set config.py Variable calibrate = False
> Then Restart speed-cam.py and monitor activity.
>
> WARNING: It is Advised to Use 320x240 Stream for Best Performance.
> Higher Resolutions Need More OpenCV Processing
> and May Reduce Data Accuracy and Reliability.
>
> Calibration Image Saved To /home/pi/speed-camera/media/images/calib-20191023-1038552.jpg
> View Calibration Image in Web Browser (Ensure webserver.py is started)
>
> ---------------------- Press cntl-c to Quit Calibration Mode -----------------------
> 2019-10-23 10:38:55 INFO speed_camera Saved media/images/calib-20191023-1038552.jpg
> 2019-10-23 10:38:55 INFO isSQLite3 Success: File is sqlite3 Format /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO db_check Success: sqlite3 Connected to DB /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO speed_camera SQL - Update sqlite3 Data in /home/pi/speed-camera/data/speed_cam.db
> 2019-10-23 10:38:55 INFO speed_camera End - R2L Ave Speed 30.3 kph Tracked 48 px in 0.192 sec Calib 150px 4408mm
> ----------------------------------------------------------------------
> 2019-10-23 10:38:55 INFO speed_camera New - 0/5 xy(75,9) Start New Track
> 2019-10-23 10:38:55 INFO speed_camera Add - 1/5 xy(65,10) 48.24 kph D=10/20 C=3 49x75=1678 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 2/5 xy(56,13) 28.11 kph D=9/20 C=4 48x70=1415 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 3/5 xy(46,13) 15.27 kph D=10/20 C=3 49x70=1521 sqpx R2L
> 2019-10-23 10:38:55 INFO speed_camera Add - 4/5 xy(55,17) 25.90 kph D=9/20 C=6 30x46=787 sqpx L2R
> 2019-10-23 10:38:55 INFO speed_camera Add - 5/5 xy(46,16) 26.50 kph D=9/20 C=4 33x47=907 sqpx R2L
> ----------------------------- Create Calibration Image -----------------------------
>
> Instructions for using media/images/calib-20191023-1038555.jpg image for camera calibration
>
> No
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#44?email_source=notifications&email_token=ABNPKZDSBWTJ6DZ3QIELIJTQQAMHBA5CNFSM4JCP7TB2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECAY5XI#issuecomment-545361629>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ABNPKZDDHVEPYGSFA7HASWDQQAMHBANCNFSM4JCP7TBQ>
> .
>
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
--
See my YouTube Channel at http://www.youtube.com/user/pageaucp
from speed-camera.
Related Issues (20)
- No Window to view camera HOT 6
- Windows not displaying on host machine HOT 1
- Clarification on calibration HOT 4
- Frames dropped due to contour out of bounds? HOT 2
- Docker Install always 11.22 HOT 1
- Calibration with different Screen Ratio and Framerate HOT 1
- No detections or speed calcs HOT 2
- No images
- How do I manually trigger the camera? HOT 4
- Can I reposition calibration marks? HOT 1
- Capture being triggered by small objects and missing all vehicles. HOT 1
- Warnings in startup log: "Unsupported V4L2 pixel format" HOT 1
- Vehicles at higher speeds being missed. HOT 4
- search-speed / Not compatible with the major rewrite? HOT 2
- Bigger resolution - `picam720` plugin - `speed_camera Out - 0/6` messages HOT 1
- Vehicles detected twice: Ideal `MO_TRACK_EVENT_COUNT`
- `MO_MIN_AREA_PX` is not the area of the rectangle
- Capture video HOT 2
- RPI 5, Bookworm - speed_cam.py process starts, then stops HOT 1
- Over reading speed on large vehicles HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from speed-camera.