[tool] [tutorial] Free AI Person Detection for Blue Iris

Hello Folks,

I am new in AI management for IP surveillance CAMs. Got runnning flawlessly AI tools with DeepStack and BI5.
But I'm having some false alerts at night with IR LEDs ON, on AI tools which are clear to be an actual alert.
As I am still testing everything I have confidence at 0-100% and Car and person detection.

false alert.png

Do I have any solution? Is there any other AI server better than Deepstack?

Thank you guys!!
 
Hello Folks,

I am new in AI management for IP surveillance CAMs. Got runnning flawlessly AI tools with DeepStack and BI5.
But I'm having some false alerts at night with IR LEDs ON, on AI tools which are clear to be an actual alert.
As I am still testing everything I have confidence at 0-100% and Car and person detection.

View attachment 63287

Do I have any solution? Is there any other AI server better than Deepstack?

Thank you guys!!
Hi, I've found the same so I record the clips in BI and use the AITool to send me telegram notifications. The tool is great in the day but I've found deepstack a bit lacking at night, even with good IR lights.
 
Last edited:
Hi
Win10
deepstack show STOP SERVER

Log AItool 1.63:
[08.06.2020, 14:17:24]: Starting analysis of C:\AIinput/AI_W1.20200608_141724362.jpg
[08.06.2020, 14:17:24]: (1/6) uploading image to DeepQuestAI Server ...
[08.06.2020, 14:17:25]: (2/6) Waiting for results ...
[08.06.2020, 14:17:25]: (3a/6) Deserializing results ...
[08.06.2020, 14:17:25]: (3b/6) Processing results:
[08.06.2020, 14:17:25]: ERROR: Processing the image C:\AIinput/AI_W1.20200608_141724362.jpg failed. Failure in AI Tool processing the image.

tried to repair with:
AItool1.65 or 1.63
install C++,
restart machine
run aitool as admin
depstack port 80 and 81

Noticed posible error:
in IE can not Access localhost:81 (in deepstack window i have message: Visit localhost to activate Deepstack) ?

Any suggestion what else to try?

LP, W
 
I solved with ports.
In Blueiris i had port 81 for web server
In Deepstack i had port 81

So changeing port to 82 in deepstack solved problems.
Then I could get to deepstack server via IE and registered with key

LP, W
 
Hi
Win10
deepstack show STOP SERVER

Log AItool 1.63:
[08.06.2020, 14:17:24]: Starting analysis of C:\AIinput/AI_W1.20200608_141724362.jpg
[08.06.2020, 14:17:24]: (1/6) uploading image to DeepQuestAI Server ...
[08.06.2020, 14:17:25]: (2/6) Waiting for results ...
[08.06.2020, 14:17:25]: (3a/6) Deserializing results ...
[08.06.2020, 14:17:25]: (3b/6) Processing results:
[08.06.2020, 14:17:25]: ERROR: Processing the image C:\AIinput/AI_W1.20200608_141724362.jpg failed. Failure in AI Tool processing the image.

tried to repair with:
AItool1.65 or 1.63
install C++,
restart machine
run aitool as admin
depstack port 80 and 81

Noticed posible error:
in IE can not Access localhost:81 (in deepstack window i have message: Visit localhost to activate Deepstack) ?

Any suggestion what else to try?

LP, W
It sounds like Deepstack needs to be activated. Have you tried browsing to (or 81) and activating Deepstack?
 
The new variable is &trigger&memo=text which allows up to 35 characters to be stored. This is great because it allows you to specify keywords ie “person” or “dog” or the actual relevant triggers?

Therefrore is it possible if you could add some variables into the program so we can do %trigger% Or similiar which will allow us to store the exact relevant trigger word? Great for high level overviews!

Ken has just added a 'memo' key to the JSON 'alertlist' command reply (update 5.29.12).

If you try to use this new key in code, just be aware that it does not appear in every alert in the JSON reply; rather only for External source alerts created via /admin?camera=x&trigger that use the &memo=text HTTP query parameter.
For a 'memo' with spaces in the text (like the example below), be sure to use %20, like &memo=my%20text.

Example...

json_alertlist_memo.png

Furthermore, when it exists, the memo text appears in the clip list (instead of 'EXTERNAL')
1591651875229.png
 
Last edited:
  • Like
Reactions: Splat and actran
I'm getting the error message "ERROR: Could not access image <link to camera file>" but it seems to occur randomly. Is this an indication on that my machine is having performance issues? Thanks
 
That sound really cool! Does the memo text show up in UI3 aswell? You can build the latest version from Github, it includes the following placeholders you can insert into the trigger url:
[detection]
[position]
[camera]
[confidence]
[detections] (all detected objects separated by comma)
[confidences] (all confidences separated by comma in the same oder as above)

So this might work: /admin?camera=x&trigger&memo=[detection]%20([confidence])

It's not yet release because I had no time to test it sufficiently and I don't want many people having problems after updating. And it includes some changes towards flagging relevant alerts (so that duplicate cams are no longer neccessary).
 
That sound really cool! Does the memo text show up in UI3 aswell? You can build the latest version from Github, it includes the following placeholders you can insert into the trigger url:
[detection]
[position]
[camera]
[confidence]
[detections] (all detected objects separated by comma)
[confidences] (all confidences separated by comma in the same oder as above)

So this might work: /admin?camera=x&trigger&memo=[detection]%20([confidence])

It's not yet release because I had no time to test it sufficiently and I don't want many people having problems after updating. And it includes some changes towards flagging relevant alerts (so that duplicate cams are no longer neccessary).

That would generate some odd results.. would give you for example 'person,car,dog 0.999999999,0.543242,0.123214'
 
That would generate some odd results.. would give you for example 'person,car,dog 0.999999999,0.543242,0.123214'

it should give "person (0.999999)". Only [detections] gives all detected&relevant objects. I agree, the confidence value could be shortened ;-)

@BrutusBoots what did you do different in your memo integration?
 
Is it possible to run this with dual stream in Blue iris? Can the trigger URL be specified which stream to trigger? Example: I set low resolution stream as main and HD as sub and have trigger URL trigger sub stream in blueiris. That would save having to duplicate camera's for SD and HD.