jon102034050
Active Member
vc1234 said:
also interested in this.
Desert_AIP said:
Simple way to view IP Cameras on TV
- Thread starter Desert_AIP
- Start date
also interested in this.
Desert_AIP
Senior Member
I have it set up in a 2x2,each at 640x480 and 20 or 30 fps.drvnbysound said:
I'll post details on the weekend.
The link to the display script I'm using has a 3x2 or 3x3 layout on it.
Desert_AIP
Senior Member
I did this in stages and added a piece at a time.
I combined several tutorials to achieve the desired effect.
First was the display cameras script.
Next was WeMo emulation.
Next was CEC command functionality.
Finally I combined the WeMo and CEC commands.
As an interim step after I got the WeMo emulation running, I used a GPIO script to drive a relay.
That way you could issue voice commands to do something like close the garage door.
What I learned from that allowed me to substitue the GPIO script for the CEC commands.
I don't think I'll post that here because I don;t wantt o cause confusion.
I'll post the commands I used with the RPi for each section. You should be able to cut and paste these in an SSH session using PuTTY or another client.
First the Display Cameras Script.
I used this tutorial:
https://community.spiceworks.com/how_to/123787-raspberry-pi-powered-surveillance-camera-monitoring-display
Start with a clean install of the latest Raspbian Jessie.
Then make sure everything is up to date
Software
Firmware
Display cameras uses Omxplayer & Screen
Install Omxplayer & Screen
Then increase the GPU memory
Use raspi-config
Set aside 256-512 MB to GPU
Also, while in raspi-config, tell the script to hold until network is available.
Select "Wait for Network at Boot" in raspi-config.
Determine the camera layout math and pixel coordinates, see link above for several examples.
Edit config.txt to set memory and display limits
You want to set the output up for display on a TV.
Add or un comment lines
For 16:9 display
sdtv_aspect=3
Group=1=CEA, mode=16=1080P 60Hz
hdmi_group=1
hdmi_mode=16
Determine your camera stream URLs. I tried this with both http and RTSP streams and it works with both.
I'm using Vivotek cameras, use the URL for your particular camera.
I also set up a view only user ID and password on the cameras so I wasn't using the root password.
The script will auto scale the streams, but I found that after it ran for awhile the screen went blank. Even with the screen saver turned off. Since I have 4 streams per camera, I ded icated one to this task and formatted it as a 640x480 native resolution and used that size for each display panel.
Create "displaycameras" script
Replace the <...camX url...> with the RTSP or RTMP or http URLs for your cameras.
NOTE: The command that calls the first camera feed (line 13) has the added option, "-b", which sets background to black
Script contents for 4x4 layout 4:3 with 640x480 sources (or windows) 21 pixel frame above and horizontally between feeds:
--------------------------------------------------------------
#!/bin/bash
### BEGIN INIT INFO
# Provides: omxplayer
# Required-Start:
# Required-Stop:
# Default-Start: 2 3 4 5
# Default-Stop: 0 1 6
# Short-Description: Displays camera feeds for monitoring
# Description:
### END INIT INFO
# Camera Feeds & Positions
top_left="screen -dmS top_left sh -c 'omxplayer --win \"0 21 640 501\" rtsp:// <...cam1_url...> -b --live '";
top_right="screen -dmS top_right sh -c 'omxplayer --win \"640 21 1280 501\" rtsp:// <...cam2_url...> --live'";
bottom_left="screen -dmS bottom_left sh -c 'omxplayer --win \"0 522 640 1002\" rtsp:// <...cam3_url...> --live'";
bottom_right="screen -dmS bottom_right sh -c 'omxplayer --win \"640 522 1280 1002\" rtsp:// <...cam4_url...> --live'";
# Camera Feed Names
# (variable names from above, separated by a space)
camera_feeds=(top_left top_right bottom_left bottom_right)
# Sleep for 1 minute on startup to allow network to establish
# This line is optional, but helps with the periodic restarts we'll initiate in crontab
# It will delay startup for one minute when you run the application.
# This may cause you to think it's not working correctly.
# I recommend commenting the line below out during testing.
sleep 1m
#---- There should be no need to edit anything below this line ----
# Start displaying camera feeds
case "$1" in
start)
for i in "${camera_feeds[@]}"
do
eval eval '$'$i
done
echo "Camera Display Started"
;;
# Stop displaying camera feeds
stop)
sudo killall omxplayer.bin
echo "Camera Display Ended"
;;
# Restart any camera feeds that have died
repair)
for i in "${camera_feeds[@]}"
do
if !(screen -list | grep -q $i)
then
eval eval '$'$i
echo "$i is now running"
fi
done
;;
*)
echo "Usage: /etc/init.d/displaycameras {start|stop|repair}"
exit 1
;;
esac
--------------------------------------------------------------
Next, change permissions for displaycameras script to make it executable
Script Use:
Start the camera feeds:
Close the camera feeds:
Re-open any closed camera feeds:
Update the rc.d file to initialize the script on startup
I recommend setting a static IP address.
This will come into play later with the WeMo emulation.
Set up automated reboot and restart using the crontab
crontab options * * * * * (five registers * = all of that category)
registers 1-5
1) min (0-59)
2) hour (0-23)
3) day of month (1-31)
4) month (1-12)
5) day of week (0-7) : 0 and 7 are both Sunday
0 3 * * * sudo reboot -> reboots the PI everyday at 3AM -> 3:00
1-59/5 * * * * /etc/init.d/displaycameras repair -> restarts the cameras every 5 minutes if they hang but not during reboots -> i.e. minute 0 to 1 not included
Edit the crontab
Add to bottom of file:
--------------------------------------------------------------
0 3 * * * sudo reboot
30 0-23/4 * * * /etc/init.d/displaycameras repair
--------------------------------------------------------------
Turn Off screen saver
BLANK_TIME=0
POWERDOWN_TIME=0
setterm -blank 0 -powerdown 0
That's it for the first part.
Here are two links I used for the next part.
Alexa WEMO
http://www.instructables.com/id/Hacking-the-Amazon-Echo/
GPIO Relay
http://www.instructables.com/id/Control-Any-Remote-From-a-Raspberry-Pi-and-Amazon-/?ALLSTEPS
I combined several tutorials to achieve the desired effect.
First was the display cameras script.
Next was WeMo emulation.
Next was CEC command functionality.
Finally I combined the WeMo and CEC commands.
As an interim step after I got the WeMo emulation running, I used a GPIO script to drive a relay.
That way you could issue voice commands to do something like close the garage door.
What I learned from that allowed me to substitue the GPIO script for the CEC commands.
I don't think I'll post that here because I don;t wantt o cause confusion.
I'll post the commands I used with the RPi for each section. You should be able to cut and paste these in an SSH session using PuTTY or another client.
First the Display Cameras Script.
I used this tutorial:
https://community.spiceworks.com/how_to/123787-raspberry-pi-powered-surveillance-camera-monitoring-display
Start with a clean install of the latest Raspbian Jessie.
Then make sure everything is up to date
Software
Code:
sudo apt-get update && sudo apt-get upgrade -yFirmware
Code:
sudo apt-get install rpi-update && echo Y | sudo rpi-updateDisplay cameras uses Omxplayer & Screen
Install Omxplayer & Screen
Code:
sudo apt-get install omxplayer
sudo apt-get install screenThen increase the GPU memory
Use raspi-config
Set aside 256-512 MB to GPU
Code:
sudo raspi-configAlso, while in raspi-config, tell the script to hold until network is available.
Select "Wait for Network at Boot" in raspi-config.
Determine the camera layout math and pixel coordinates, see link above for several examples.
Edit config.txt to set memory and display limits
You want to set the output up for display on a TV.
Code:
sudo nano /boot/config.txt Add or un comment lines
For 16:9 display
sdtv_aspect=3
Group=1=CEA, mode=16=1080P 60Hz
hdmi_group=1
hdmi_mode=16
Determine your camera stream URLs. I tried this with both http and RTSP streams and it works with both.
I'm using Vivotek cameras, use the URL for your particular camera.
I also set up a view only user ID and password on the cameras so I wasn't using the root password.
The script will auto scale the streams, but I found that after it ran for awhile the screen went blank. Even with the screen saver turned off. Since I have 4 streams per camera, I ded icated one to this task and formatted it as a 640x480 native resolution and used that size for each display panel.
Create "displaycameras" script
Replace the <...camX url...> with the RTSP or RTMP or http URLs for your cameras.
NOTE: The command that calls the first camera feed (line 13) has the added option, "-b", which sets background to black
Code:
sudo nano /etc/init.d/displaycamerasScript contents for 4x4 layout 4:3 with 640x480 sources (or windows) 21 pixel frame above and horizontally between feeds:
--------------------------------------------------------------
#!/bin/bash
### BEGIN INIT INFO
# Provides: omxplayer
# Required-Start:
# Required-Stop:
# Default-Start: 2 3 4 5
# Default-Stop: 0 1 6
# Short-Description: Displays camera feeds for monitoring
# Description:
### END INIT INFO
# Camera Feeds & Positions
top_left="screen -dmS top_left sh -c 'omxplayer --win \"0 21 640 501\" rtsp:// <...cam1_url...> -b --live '";
top_right="screen -dmS top_right sh -c 'omxplayer --win \"640 21 1280 501\" rtsp:// <...cam2_url...> --live'";
bottom_left="screen -dmS bottom_left sh -c 'omxplayer --win \"0 522 640 1002\" rtsp:// <...cam3_url...> --live'";
bottom_right="screen -dmS bottom_right sh -c 'omxplayer --win \"640 522 1280 1002\" rtsp:// <...cam4_url...> --live'";
# Camera Feed Names
# (variable names from above, separated by a space)
camera_feeds=(top_left top_right bottom_left bottom_right)
# Sleep for 1 minute on startup to allow network to establish
# This line is optional, but helps with the periodic restarts we'll initiate in crontab
# It will delay startup for one minute when you run the application.
# This may cause you to think it's not working correctly.
# I recommend commenting the line below out during testing.
sleep 1m
#---- There should be no need to edit anything below this line ----
# Start displaying camera feeds
case "$1" in
start)
for i in "${camera_feeds[@]}"
do
eval eval '$'$i
done
echo "Camera Display Started"
;;
# Stop displaying camera feeds
stop)
sudo killall omxplayer.bin
echo "Camera Display Ended"
;;
# Restart any camera feeds that have died
repair)
for i in "${camera_feeds[@]}"
do
if !(screen -list | grep -q $i)
then
eval eval '$'$i
echo "$i is now running"
fi
done
;;
*)
echo "Usage: /etc/init.d/displaycameras {start|stop|repair}"
exit 1
;;
esac
--------------------------------------------------------------
Next, change permissions for displaycameras script to make it executable
Code:
sudo chmod 755 /etc/init.d/displaycamerasScript Use:
Start the camera feeds:
Code:
/etc/init.d/displaycameras startClose the camera feeds:
Code:
/etc/init.d/displaycameras stopRe-open any closed camera feeds:
Code:
/etc/init.d/displaycameras repairUpdate the rc.d file to initialize the script on startup
Code:
sudo update-rc.d displaycameras defaults
sudo update-rc.d displaycameras remove I recommend setting a static IP address.
This will come into play later with the WeMo emulation.
Set up automated reboot and restart using the crontab
crontab options * * * * * (five registers * = all of that category)
registers 1-5
1) min (0-59)
2) hour (0-23)
3) day of month (1-31)
4) month (1-12)
5) day of week (0-7) : 0 and 7 are both Sunday
0 3 * * * sudo reboot -> reboots the PI everyday at 3AM -> 3:00
1-59/5 * * * * /etc/init.d/displaycameras repair -> restarts the cameras every 5 minutes if they hang but not during reboots -> i.e. minute 0 to 1 not included
Edit the crontab
Code:
crontab -eAdd to bottom of file:
--------------------------------------------------------------
0 3 * * * sudo reboot
30 0-23/4 * * * /etc/init.d/displaycameras repair
--------------------------------------------------------------
Turn Off screen saver
Code:
sudo nano /etc/kbd/config BLANK_TIME=0
POWERDOWN_TIME=0
Code:
sudo nano ~/.bashrcsetterm -blank 0 -powerdown 0
That's it for the first part.
Here are two links I used for the next part.
Alexa WEMO
http://www.instructables.com/id/Hacking-the-Amazon-Echo/
GPIO Relay
http://www.instructables.com/id/Control-Any-Remote-From-a-Raspberry-Pi-and-Amazon-/?ALLSTEPS
Thanks for sharing, very interesting.
Are you communicating with the cameras over the wire or WiFi ?
With my cameras, I have a main stream (1920x1080) and one sub-stream (640x480, I think). The main streams go to an Xprotect Milestone box, and then you can view live/recording by connecting to the Milestone server or directly to the cameras. There seems to be no problem using the same main stream both for recording to the Xprotect box and viewing using a web browser connected to the camera.
Yet, I cannot figure out a simple way to display permanently a hi-res 4x4 image via an HDMI input on my TV.
I thought about running an Xprotect app or maybe just a browser on an Intel Windows stick connected to the TV HDMI port, but I am not sure it's a good idea since each main stream uses about 2-3Mbit/s, and I'd prefer to use WiFi rather than a pull an rj45 cable for something which is mere a convenience -- I can easily see the same 4x4 on my iPad. Also, I am not sure if a 640x480 4x4 on a big screen TV would make any sense in comparison to the full resolution display on the iPad.
Are you communicating with the cameras over the wire or WiFi ?
With my cameras, I have a main stream (1920x1080) and one sub-stream (640x480, I think). The main streams go to an Xprotect Milestone box, and then you can view live/recording by connecting to the Milestone server or directly to the cameras. There seems to be no problem using the same main stream both for recording to the Xprotect box and viewing using a web browser connected to the camera.
Yet, I cannot figure out a simple way to display permanently a hi-res 4x4 image via an HDMI input on my TV.
I thought about running an Xprotect app or maybe just a browser on an Intel Windows stick connected to the TV HDMI port, but I am not sure it's a good idea since each main stream uses about 2-3Mbit/s, and I'd prefer to use WiFi rather than a pull an rj45 cable for something which is mere a convenience -- I can easily see the same 4x4 on my iPad. Also, I am not sure if a 640x480 4x4 on a big screen TV would make any sense in comparison to the full resolution display on the iPad.
The limiting factors are going to be your network and camera (how many concurrent streams they can support and how much bandwidth you can support)vc1234 said:
Most commonly you're going to have 2 streams going to your recording device, the compressed H264 (most commonplace these days, but could be MP4V, etc) and generally a MJPEG low res, low FPS stream that would be used for motion based detection or analytics....that's how everyone is doing it these days. You're not grabbing an additional stream from the camera, you are grabbing the stream from the VMS in your case.
The most simple way to get the video to the monitor is a small client which is outside of any VMS, and grabbing a different stream from the camera than what is going to be used on the VMS side, typically ONVIF is the easiest to do this with.
Desert_AIP
Senior Member
All of my cameras are PoE hardwired.
The RPi is on a PoE hardline connection too.
My entire LAN is gigabit, but the PoE devices are limited to megabit.
The RPi is on a PoE hardline connection too.
My entire LAN is gigabit, but the PoE devices are limited to megabit.
Yes, that's what I thought too. Just cannot find something small to play the stream with. I heard about the Inter Windows stick that plugs into the HDMI port, but do not know if it's any good. Comments on Amazon are not encouraging. Due to esthetics, even an RPI would probably look out of place. The TV is mounted on a brick wall so it's not easy to add any wires beyond what's already there.DELInstallations said:
Desert_AIP
Senior Member
You could use one of those Android boxes and an IPCamera app.
pete_c
Guru
Thank you Desert_AIP for providing the camera stuff for the RPi.
I have a few of the RPi's and have not really attached them to an video stuff.
Tried RPi2 KODI a while ago and it was slower than my currently utilized Aopen Digital engines so left it alone.
I do stream CCTV to Kodi and it works with favorites.
So above you are doing a 4 X 4 camera stream (640X480) to your LCD TV (1900X1280?)
I have a few of the RPi's and have not really attached them to an video stuff.
Tried RPi2 KODI a while ago and it was slower than my currently utilized Aopen Digital engines so left it alone.
I do stream CCTV to Kodi and it works with favorites.
So above you are doing a 4 X 4 camera stream (640X480) to your LCD TV (1900X1280?)
There's a mobile app for both Android and iPhone from Milestone which works OK. Would the Android phone/tablet app run on an Android HDMI stick just as well ? E.g. on this one:Desert_AIP said:
http://www.htpcbeginner.com/cx-919-android-stick/
I do not have much experience with Android OS, so I am not sure if sticks use the same OS as ordinary phones.
Yes, it is a possibility -- the box can be hidden behind the TV, but I am still attracted to the idea of plugging in an Android stick, also behind the TV, into an HDMI port. The stick can also be used for other purposes. I guess I'll get an Amazon Fire TV stick and try to install the Milestone mobile client and see how good it'll look on the screen.DELInstallations said:
Desert_AIP
Senior Member
The WeMo emulation setup is pretty straightforward, just a few commands.
Let each one run until completion before moving on.
Some take several minutes.
Then navigate to the echo-master directory and start the python script.
After you start the script you have to tell Alexa to discover your devices.
Say
She will be busy for about 30 seconds.
When she's done she should report she found one device (unless you have others).
Then you can test the emulation.
Say
She should respond "OK" after she gets the proper response back from the RPi, and on the RPi screen you'll see several lines describing that "device" was turned on or off.
Take a look at the second link for adding and controlling a GPIO relay to the WeMo emulation.
I used that and modified it to send the CEC commands.
Also note, you can add multiple WeMo devices to a single RPi with one IP address. Each device simply needs an individual port number. The notes in the WeMo link above explains this a lot.
I found I can give a device multiple names, each with a different port number, then group them int he action script so a command to any of them has the same effect.
So I may use the term "Flood Lights", "Security Lights" and "Exterior Floods" all to describe my perimeter lights.
If I say "Alexa, turn on Flood Lights" or "Alexa, turn on Exterior Floods", etc. she sends the on command to the appropriate device. This saves programming lines in the Python script, and does not tie you to a specific name for a unit.
(I'm still working on sending out UPB commands BTW, I cant get the serial commands to be sent to my PIM).
Also, the Echo is fairly smart at parsing commands.
So "Alexa, turn on Security Lights", and "Alexa, Security Lights on." give similar results.
To add multiple devices give each a unique name and port number in the TRIGGERS line.
i.e.
TRIGGERS = {"Garage": 52000,"Security Lights": 52001,"Security": 52002,"Bath": 52003}
Let each one run until completion before moving on.
Some take several minutes.
Code:
sudo apt-get install python-pip ; sudo pip install requests
sudo wget "https://github.com/toddmedema/echo/archive/master.zip"
sudo unzip master.zip -d echoThen navigate to the echo-master directory and start the python script.
Code:
cd echo/echo-master
python example-minimal.pyAfter you start the script you have to tell Alexa to discover your devices.
Say
Code:
"Alexa, discover my devices."She will be busy for about 30 seconds.
When she's done she should report she found one device (unless you have others).
Then you can test the emulation.
Say
Code:
"Alexa, turn my device on (or off)."She should respond "OK" after she gets the proper response back from the RPi, and on the RPi screen you'll see several lines describing that "device" was turned on or off.
Take a look at the second link for adding and controlling a GPIO relay to the WeMo emulation.
I used that and modified it to send the CEC commands.
Also note, you can add multiple WeMo devices to a single RPi with one IP address. Each device simply needs an individual port number. The notes in the WeMo link above explains this a lot.
I found I can give a device multiple names, each with a different port number, then group them int he action script so a command to any of them has the same effect.
So I may use the term "Flood Lights", "Security Lights" and "Exterior Floods" all to describe my perimeter lights.
If I say "Alexa, turn on Flood Lights" or "Alexa, turn on Exterior Floods", etc. she sends the on command to the appropriate device. This saves programming lines in the Python script, and does not tie you to a specific name for a unit.
(I'm still working on sending out UPB commands BTW, I cant get the serial commands to be sent to my PIM).
Also, the Echo is fairly smart at parsing commands.
So "Alexa, turn on Security Lights", and "Alexa, Security Lights on." give similar results.
To add multiple devices give each a unique name and port number in the TRIGGERS line.
i.e.
TRIGGERS = {"Garage": 52000,"Security Lights": 52001,"Security": 52002,"Bath": 52003}
bucko
Active Member
I solved this issue simply and cheaply as well. Just use a TV Box ( I use a Minix Box). Load the IP Cam Pro app on it and you have nice HD cam viewing, control, sound, playback, recording,
Plus the assed bonus of Kodi bringing you NO Cost worldwide TV and Movie viewing.
Problem solved >$100
Plus the assed bonus of Kodi bringing you NO Cost worldwide TV and Movie viewing.
Problem solved >$100
