I dont see anything in the 0.103 release notes which would have affected this
Funny even folder_watch does not work. I did spin another ha instance and was able to make folder_watch work in the first try. In my current ha installation it does not work, I wonder if any plugin or extensions is getting in the way…not related to this thread but just following up to my previous question.
You should increase the confidence threshold
It’s on 85% already and actually some time missing the person detection because of that threshold (some people detected around 75% confidence)…
Getting into trying to set this up with HACS, and getting this error:
`Platform error image_processing.deepstack_object - No module named 'homeassistant.util.pil``
What am I missing?
Update to latest release
Hi @robmarkcole, thanks for your integration!
I think I followed all the steps, tried both deepstack and deepstack:noavx (even if my cpu flags shows avx2) but I still get this error in HA logs:
Depstack error : Error from Deepstack: Detection endpoint not activated
I’m running HA on docker (not hassio) and deepstack as a docker image as well. Activated with no issues and it shows 729 days remaining.
In home assistant I defined a local_file camera, and I feed that entity into the image_processing component, but when I trigger the image_processing.scan service I get the aforementioned error.
What am I missing?
Try restarting the container. Otherwisd its a deepstack issue for their forums
Hi everyone!
I got some problem to use object recognation with my camera
It’s working with a jpg camera:
- platform: local_file
file_path: /tmp/image.jpg
So my problem is my real camera, setup like this:
- platform: mjpeg
mjpeg_url: http://192.168.0.xx:2971/camera/stream
name: Honor8_cuisine
State is always Unknown with my real camera
SO, what type of camera do i need to use in HA ?
This morning, it’s working! i don’t know what really happens!
I have just setting this up and get the following error, i have the dependencies installed.
Hassio V 0.103.6
2020-01-11 18:19:51 WARNING (MainThread) [homeassistant.loader] You are using a custom integration for deepstack_object which has not been tested by Home Assistant. This component might cause stability problems, be sure to disable it if you do experience issues with Home Assistant.
2020-01-11 18:19:51 ERROR (MainThread) [homeassistant.config] Platform error: image_processing
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/config.py", line 761, in async_process_component_config
platform = p_integration.get_platform(domain)
File "/usr/src/homeassistant/homeassistant/loader.py", line 232, in get_platform
f"{self.pkg_path}.{platform_name}"
File "/usr/local/lib/python3.7/importlib/__init__.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1006, in _gcd_import
File "<frozen importlib._bootstrap>", line 983, in _find_and_load
File "<frozen importlib._bootstrap>", line 967, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 677, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 728, in exec_module
File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
File "/config/custom_components/deepstack_object/image_processing.py", line 23, in <module>
from homeassistant.util.pil import draw_box
ModuleNotFoundError: No module named 'homeassistant.util.pil'
Any help appreciated.
Update HA to latest version
1 Like
All good thanks was me being a bit hesitant at updating should of been the first thing I did.
1 Like
Hi
I have both face and. vision sion enabled in a single docker (using multiple -e option).
face detection works. Object and face reco does not.
I have triple checked folders whitelisting and they are.
In the logs of docker i see nothing happening (even though face detection works)
I have attempted multiple times to register faces , i call the service but no feedback and no matches.
Is there somewhere inthe docker a log i can check? i can login by connecting through portainerconsole to the deepstack container.
Tx in advance
Do I need to use the google coral HASS-Google-Coral integration for google coral device, or I can use the https://github.com/robmarkcole/HASS-Deepstack-object with it?
Thx!
Thx, this is what I use now, but I don’t see any file saved to the folder so I thought I am not using the right integration.
I am trying to debug it …
UPDATE:
I think the reason that I don’t see any image saved, is that I see person == 0, and in that case image is not saved.
Does the integration support any target object? currently, as I understand, I need to list the targets I want and person is the default. Any way to support something like:
target: *
Is it is possible at all to define list of targets (from the code I don’t see it)? something like:
target: clock, bench, car
Thx
An image_processing.object_detected event is fired for each object, you should use that.
Previously more data was exposed via attributes but it was very messy approach an unsatisfactory so removed