CameraEvents Service - Dahua Events to mqtt

Hi,

my Model is the Synology DS718+ ( LINK to Specs ).

I am using the Synology Docker Package and for example this Image: Mosquitto and this Unifi Controller. Both are running without any Problem.

Here is also a list of the architecture per Synology NAS: LINK
There is stated that the Docker Package uses X64, afaik.

Thanks for your help.

Regards Sven
 
Last edited:
psycik said:
As per: NodeJS Module: node-dahua-api

psyciknz/CameraEvents

This is a python service I run in docker. One of the first I wrote from scratch, so the documentation is probably a little light.

I've updated the readme for some run information, (git clone, python CameraEvents.py). And will figure out making the docker image automatically (this is all learning to me).

The main things it can do are to convert all VideoMotion events and IVS events into switches that can then be read by OpenHAB and HomeAssistant (or anything else that uses MQTT). IT also has the ability to pull a snapshot from the camera and post that as binary data to mqtt.
Click to expand...

Hi Psyciknz.

I'm also in New Zealand. I'm a bit stuck trying to use your github.

Whats the command for building the docker image. I'm used to having a file called "dockerbuild" not "dockerbuild.cross". I can't figure out how to use it the .cross file. PS. I'm using linux.

If i try and just run the script directly, i get the following error:

mike@server:~/.docker/config/dahua-mqtt-script$ python3 CameraEvents.py
File "CameraEvents.py", line 282
except Exception,ivsExcept:
^
SyntaxError: invalid syntax
 
Last edited:
Hi,

I finally got it running in docker on my synology NAS.

My camera is a IPC-HDW5231R-ZE. I created a rule on the camera for CrossLineDetection, which is detected by the script correctly. But there are not Details on the IVS data like e.g. crossingdirection. Is IVS data depending on the camera model or is there something more to define in the config?

Here is the log for one event:
Code: 
2019-06-23 13:11:54,254 - __main__ - DEBUG - [Carport]: --myboundary
Content-Type: text/plain
Content-Length:44
Code=CrossLineDetection;action=Start;index=0

2019-06-23 13:11:54,254 - __main__ - ERROR - Error getting IVS data: 'data'
2019-06-23 13:11:54,306 - __main__ - INFO - Snapshot Url: http://172.17.2.20:80/cgi-bin/snapshot.cgi?channel=0
2019-06-23 13:11:56,681 - __main__ - DEBUG - [Carport]: --myboundary
Content-Type: text/plain
Content-Length:43
Code=CrossLineDetection;action=Stop;index=0

Regards, Sven
 
Last edited:
Ole said:
Hi,

I finally got it running in docker on my synology NAS.

My camera is a IPC-HDW5231R-ZE. I created a rule on the camera for CrossLineDetection, which is detected by the script correctly. But there are not Details on the IVS data like e.g. crossingdirection. Is IVS data depending on the camera model or is there something more to define in the config?

Here is the log for one event:
Code: 
2019-06-23 13:11:54,254 - __main__ - DEBUG - [Carport]: --myboundary
Content-Type: text/plain
Content-Length:44
Code=CrossLineDetection;action=Start;index=0

2019-06-23 13:11:54,254 - __main__ - ERROR - Error getting IVS data: 'data'
2019-06-23 13:11:54,306 - __main__ - INFO - Snapshot Url: http://172.17.2.20:80/cgi-bin/snapshot.cgi?channel=0
2019-06-23 13:11:56,681 - __main__ - DEBUG - [Carport]: --myboundary
Content-Type: text/plain
Content-Length:43
Code=CrossLineDetection;action=Stop;index=0

Regards, Sven
Click to expand...

The lack of details suggests your camera it not a high enough firmware, or just doesn’t post the json that I saw in mine.

Do you get an event fired? As it should just end up like a video motion event which also has very little details.
 
Hi,

yes I get an event. But the payload is just "CrossLineDetection". For MotionDetection there are two messages with payload "on" and "off".

Details of my camera:
Code: 
Device Type:      IPC-HDW5231R-ZE
System Version:   2.622.0000000.18.R, Build Date: 2017-11-10
WEB Version:      3.2.1.515090
ONVIF Version:    16.12(V2.4.1.510971)

Could you check with your camera what version you are running?
Is there a way I can check if my camera provides a different json string?

EDIT:
Updated the Firmware:
Code: 
Device Type:     IPC-HDW5231R-ZE
System Version:   V2.800.0000005.0.R, Build Date: 2019-03-25
WEB Version:   V3.2.1.709882
ONVIF Version:   16.12(V2.4.3.651299)
Now I get more Details :-)
"CrossLineDetection With Human in LeftToRight direction for Entrance Region"


Regards Sven
 
Last edited:
Ole said:
Hi,

yes I get an event. But the payload is just "CrossLineDetection". For MotionDetection there are two messages with payload "on" and "off".

Details of my camera:
Code: 
Device Type:      IPC-HDW5231R-ZE
System Version:   2.622.0000000.18.R, Build Date: 2017-11-10
WEB Version:      3.2.1.515090
ONVIF Version:    16.12(V2.4.1.510971)

Could you check with your camera what version you are running?
Is there a way I can check if my camera provides a different json string?

EDIT:
Updated the Firmware:
Code: 
Device Type:     IPC-HDW5231R-ZE
System Version:   V2.800.0000005.0.R, Build Date: 2019-03-25
WEB Version:   V3.2.1.709882
ONVIF Version:   16.12(V2.4.3.651299)
Now I get more Details :)
"CrossLineDetection With Human in LeftToRight direction for Entrance Region"


Regards Sven
Click to expand...

That's great. I might, cos it annoys me, make the detected "things" filterable. ie I get "smoke" when it's fuzziness on the camera, so I will exclude those. That will just leave Human and Vehicle.
 
Hi I still have not been able to get this going.

This is my docker-compose.yaml

Code: 
  dahua_camera_events:

    container_name: dahua_camera_events
    image: psyciknz/cameraevents
    restart: always

I was getting this error in the log:

Code: 
2019-06-26 09:49:18,243 - __main__ - ERROR - Error starting:Failed to open/find all files,
2019-06-26 09:49:18,242 - __main__ - INFO - Loading config,

So i went and created a config.ini using config-master.ini and changed the docker-compose to:

Code: 
  dahua_camera_events:
    container_name: dahua_camera_events
    image: psyciknz/cameraevents
    restart: always
    volumes:
      - ${DOCKERCONFDIR}/dahua_camera_events/config.ini:/config.ini

However I still get errors in the log for the dahua_camera_events container

Code: 
2019-06-26 09:48:51,381 - __main__ - ERROR - Error starting:Failed to open/find all files,
2019-06-26 09:48:51,381 - __main__ - INFO - Loading config

Q: Is there a different settings file I need to use for the docker image vs the github page?

Q: Can someone please post their compose.yaml ?
 
Last edited:
kiwijunglist said:
Hi I still have not been able to get this going.

This is my docker-compose.yaml

Code: 
  dahua_camera_events:

    container_name: dahua_camera_events
    image: psyciknz/cameraevents
    restart: always

I was getting this error in the log:

Code: 
2019-06-26 09:49:18,243 - __main__ - ERROR - Error starting:Failed to open/find all files,
2019-06-26 09:49:18,242 - __main__ - INFO - Loading config,

So i went and created a config.ini using config-master.ini and changed the docker-compose to:

Code: 
  dahua_camera_events:
    container_name: dahua_camera_events
    image: psyciknz/cameraevents
    restart: always
    volumes:
      - ${DOCKERCONFDIR}/dahua_camera_events/config.ini:/config.ini

However I still get errors in the log for the dahua_camera_events container

Code: 
2019-06-26 09:48:51,381 - __main__ - ERROR - Error starting:Failed to open/find all files,
2019-06-26 09:48:51,381 - __main__ - INFO - Loading config

Q: Is there a different settings file I need to use for the docker image vs the github page?

Q: Can someone please post their compose.yaml ?
Click to expand...
  1. cameraevents:
  2. container_name: cameraevents
  3. depends_on:
  4. - mosquitto
  5. environment:
  6. - ARCH=amd64
  7. - BASEPATH=/hdd/docker-data
  8. #env_file:
  9. # - /hdd/docker-data/env/arch.env
  10. image: psyciknz/cameraevents
  11. restart: always
  12. networks:
  13. - ha
  14. volumes:
  15. - ${BASEPATH}/ha-shared/cameraevents/:/opt/cameraevents/conf
  16. - "/etc/timezone:/etc/timezone:ro"
  17. - "/etc/localtime:/etc/localtime

Sorry posted from phone. Volume command is just to the directory.

The instructions are a little light as I wrote it for me then decided to publish. And I have have much feed back either way to the problems with it.
 
Just the config.ini

The problem with most docker containers is they can’t prepopulate the host volumes with a default config. So always have to revert to this sort of manual config to start with
 
I've just updated the readme and added run instructions.

I probably wouldn't use my entry of ${BASEPATH}/ha-shared/cameraevents/ since there's a variable in there and it defines a fixed path.
But if you make a cameravents dir, put the config in there, then run the
Code: 
docker run -v ${PWD}:/opt/cameraevents/conf psyciknz/cameraevents
you should hopefully get something going.
 
  • Like
Reactions: kiwijunglist
It's working.

I haven't played around with all the events yet, so I'm not sure what's supported by the docker.

Can the docker send different mqtt messages for different events so I can have a sensor for face detection in home assistant as well as for motion?
 
kiwijunglist said:
It's working.

I haven't played around with all the events yet, so I'm not sure what's supported by the docker.

Can the docker send different mqtt messages for different events so I can have a sensor for face detection in home assistant as well as for motion?
Click to expand...
Not sure. POSs not. They might all just get identified as Ivs. Can you filter on the content is they’re all in the same topic?

My camera doesn’t tend to do face recognition so I haven’t played with it. I just use line detection and trip wire (and video motion)
 
@psycik Do you know what units the Center and Bounding Box values are?
The values I receive appear to be too large to be pixels.
I was hoping to use this information to locate where people/cars etc are and take actions accordingly.
 
You must log in or register to reply here.
Top Bottom