Who uses Dahua AI capable cameras? Reliable AI for triggering events? Pro's/con's?

TBurt said:
I am impressed with the accuracy from the onboard AI but with regards to BI recording the triggers it seems to vary wildly. What I mean is when you click the image to load the video of the event BI+ DS usually takes you to the begins of the trigger. With the cam AI Blue Iris seems to randomly pick where to take you on clicking it. Sometimes the car is just coming into view, sometimes it is just leaving the cameras field of view. While other times it shows nothing. I have tried using BI and cam AI alone. I've tried it with BI+DS to tag/label the motion object that was triggered. That at least tells me it was a car with 89% for example. But again BI and DS's clip that it found sometimes does not show anything, like what I am seeing when I actually view the clips. I have to manually wind it back to see the trigger. It captures the trigger every time though. I tried every permutation and combination of IVS with car/person/intrusion/trip wire/appears/enters/etc but I get the same results. So, I then tested SMD with human and motor vehicle. I made motion zones for just the areas I was interested in and told it to look for people/cars in those areas. I get the same thing in Blue Iris. I must be screwing something up configuration wise in BI. Or does BI just normally act this way with ONVIF triggers in general? BI does not act this way when it controls the triggers. I played with pre-trigger buffers, but I have come to the conclusion the buffers only work with BI motion sensing. I wish it would allow me to choose to start showing me clips say 1.2 seconds into the triggered recordings or something like that. If I had to check for a car that drove by my house in the last 24 hours and had to click on hundreds of AI motion triggered clips and then had to manually go back and forward to find the actual triggered object (car/person) I would lose what little sanity I have left. Now imagine you were trying to find a trigger event that happened sometime during the last week... OMG. This setup of BI simply recording AI triggers from the camera is bordering on useless. No one would want to do that. I would rather go by a Dahua NVR with AI than this. Money is tight for me as is with all the stuff going on in the world today. Hell, I would rather buy a Reolink NVR with AI and Reolink cameras than this... ok, maybe I would not go that far! Haha I must be missing something. Sure, it captures and records nearly 99.999% of motion objects triggers correctly but conversely it lacks nearly 99.999% user friendliness to go along with it. Sure I could make mutltiple cloans to help with finding events from the AI camera, but then I am having to make up things to fix this "fix". On a side note, BI+DS last night got all the cars and people also. I was hoping this setup would take me to another level in motion object detection accuracy. Instead, it takes me to another level of making band aids in BI to overcome a shortcoming of this setup. IE, make a cloan and have it do X and Y, etc. How is this any better than the standard BI? I am running one of the latest BI versions, if not the latest. Maybe being on the bleeding edge of BI technology has finally come back to bite me? Do I need to go back to an older more stable version? If so, what version are you using? Or am I just missing the bigger picture and I am worrying about a non-issue? Look i have done it again... made a long meandering post. I have a massive headache. I need to take two ibuprofens and have a nap. Any ideas, recommendations?
Click to expand...
By the way, the issue you raised with regards to the alerts not aligning with the pre-trigger buffer when using A.I. capable cameras (those that are triggered via External or ONVIF in B.I.) was recently resolved as of Blue Iris 5.5.5.10.
 
  • Like
Reactions: djernie
jrbeddow said:
By the way, the issue you raised with regards to the alerts not aligning with the pre-trigger buffer when using A.I. capable cameras (those that are triggered via External or ONVIF in B.I.) was recently resolved as of Blue Iris 5.5.5.10.
Click to expand...

I noticed that. It had already started working within reason for me just changing the version a while back. It was not placing my starting point for videos all over the place and at least putting it in the middle of the screen. So it was easy enough to find the spot of trigger. It was better, though not the way I would expect it to work. When I installed the 5.5.5.10 it was as if BI handled ONVIF triggering from my AI cameras just like BI handled its own motion triggering and pre-buffering. I tried it with both Dahua AI, and Hikvision AI ONVIF triggers. Both are working wonderfully. Bravo!
 
  • Like
Reactions: djernie and jrbeddow
Hi guys ,
I've read the full five pages of this string and have learned a lot. thank you all for your contributions. this is a great resource when using (dahua) camera analytics.

I have a couple of related questions:

-- when using camera analytics as described: is it possible to get different alert categories in BI to indicate the type of analytic event that was triggered, or, are they all essentially generic events within BI tagging /filtering?

-- are the bounding boxes from the camera alert picked up by BI so we see those on the recording/alert?

Thanks again for the great info
 
  • Like
Reactions: Yo Momma
CalebCC said:
Hi guys ,
I've read the full five pages of this string and have learned a lot. thank you all for your contributions. this is a great resource when using (dahua) camera analytics.

I have a couple of related questions:

-- when using camera analytics as described: is it possible to get different alert categories in BI to indicate the type of analytic event that was triggered, or, are they all essentially generic events within BI tagging /filtering?

-- are the bounding boxes from the camera alert picked up by BI so we see those on the recording/alert?

Thanks again for the great info
Click to expand...

Unfortunately at the moment all BI gets is that the camera triggered based on your setup in the camera.

It does not show the box around the object in BI.

Now if you want it to have that capability, send an email to BI support asking for it (I have and if others do it might be a possibility) OR use Deepstack to then take the camera triggered event and do that within BI.
 
  • Like
Reactions: Yo Momma and TBurt
wittaj said:
Unfortunately at the moment all BI gets is that the camera triggered based on your setup in the camera.

It does not show the box around the object in BI.

Now if you want it to have that capability, send an email to BI support asking for it (I have and if others do it might be a possibility) OR use Deepstack to then take the camera triggered event and do that within BI.
Click to expand...

Thanks. I will do both of your suggestions ( the support request now and testing with deep later, once I get into my configuration)

getting those native features functioning would certainly enhance BI. The workflow you guys have described of relying on camera analytics and outsourcing all that processing with the extreme high level of accuracy sure make sense. One would think the BI guys would embrace it.

as others have suggested, possibly adding deep stack on top of that might be better still ( because there would be so few false alarms sent to deepstack the cpu processing would be minimal). however if there was native integration with alert /bookmark filters and bounding boxes one wouldn't need the additional step.
 
I am sure it is something coming in a future revision. As long as the ONVIF tags the trigger as human or vehicle, the developer should be able to do it. But until enough people request it, it is something on the backburner.

I use a few of my cams AI along with DS and as you stated, the resources is minimal at that point for DeepStack as the camera has already identified it as movement of interest.
 
  • Like
Reactions: Yo Momma, TBurt and djernie
wittaj said:
Unfortunately at the moment all BI gets is that the camera triggered based on your setup in the camera.

It does not show the box around the object in BI.

Now if you want it to have that capability, send an email to BI support asking for it (I have and if others do it might be a possibility) OR use Deepstack to then take the camera triggered event and do that within BI.
Click to expand...

The closest thing to that would be to send the alert triggered by the camera to Deep Stack and have it
CalebCC said:
Hi guys ,
I've read the full five pages of this string and have learned a lot. thank you all for your contributions. this is a great resource when using (dahua) camera analytics.

I have a couple of related questions:

-- when using camera analytics as described: is it possible to get different alert categories in BI to indicate the type of analytic event that was triggered, or, are they all essentially generic events within BI tagging /filtering?

-- are the bounding boxes from the camera alert picked up by BI so we see those on the recording/alert?

Thanks again for the great info
Click to expand...

The closest thing to that would be to send the alert triggered by the camera to Deep Stack and have it categorize the alerts. This is actually a little better as DS will label a vehicle for example a truck or a bus... whereas the AI from Dahua cameras (or even Hikvision) is just a generic vehicle. You do lose if the trigger was an Intrusion, Tripwire, or SMD though. Blue Iris will just label it as "External" or "Onvif" + Car/truck/person/whatever. To email, BI to ask them to add further functionality you would need to get the Onvif information sent by the camera(s). Onvif Manager will give you all the information they would need.
 
  • Like
Reactions: Yo Momma
Has Dahua added restrictions in the firmware for the number of trip wires in IVS that weren't there before?

In my 4Kt's It says 3 lines max! I'm sure I've seen previous postings where people have drawn lots and lots of trip wires to ensure detection. Also, not sure if it's a firmware issue, running 120.000...14R but after the 1st one is drawn, I can't seem to get it to allow me to draw more. Sometimes I get a 2nd trip wire up, but then it won't let me click to deposit the anchors....
 
They have always had a limit, I think 7 or 8.

It takes a lot of processing power for it to work.

Many of us wouldn't use more than 2 or 3 because we have seen that too many and the processor is maxed out and then motion is missed.

So they probably capped this one to ensure it is able to keep up.
 
  • Like
Reactions: CCTVCam
I wonder, does my zig zag continuous single line count as 1 ?
 
  • Like
Reactions: Mark_M
I don't know what's going on here. I went into my camera this time using the both direction setting, and it let me draw loads of lines, in fact when I wanted to stop, it wouldn't let me and made me draw more!!!!

Getting really confused. I'll start a different thread as I clearly need some help.
 
  • Like
Reactions: samplenhold, bigredfish and looney2ns
Thread for 2021.... but oh well.
Here's my mini PTZ purchased from Andy in 2021.

Impressive it could detect the human behind a fence/bush while the dog was not considered
1681332618331.png
 
  • Like
Reactions: CanCuba and bigredfish
  • Like
Reactions: looney2ns, Mark_M and bigredfish
Wow, that is amazing. Can you share that with Blue Iris for facial recognition? Also, I see the cameras I purchased as Face Detection but not people detection. Does that mean it only does face or will do both? I purchased the famed 5442, and I did not read it does cars, so can I keep AI for cars?
 
mbh said:
Wow, that is amazing. Can you share that with Blue Iris for facial recognition? Also, I see the cameras I purchased as Face Detection but not people detection. Does that mean it only does face or will do both? I purchased the famed 5442, and I did not read it does cars, so can I keep AI for cars?
Click to expand...

I'm making the move to BI later this year so will be able to better understand linking BI and CodeProject for facial detection and recognition.

The 5442 has AI which detects which objects are cars and people. This applies the IVS rules which makes it much easier to reduce false detections (leaves, aninmals, etc) from people and vehicles. The 5442 also does face detection as the cameras I've shown do but you can't run IVS and face detection at the same time.
 
  • Like
Reactions: looney2ns and bigredfish
CanCuba said:
I'm making the move to BI later this year so will be able to better understand linking BI and CodeProject for facial detection and recognition.

The 5442 has AI which detects which objects are cars and people. This applies the IVS rules which makes it much easier to reduce false detections (leaves, aninmals, etc) from people and vehicles. The 5442 also does face detection as the cameras I've shown do but you can't run IVS and face detection at the same time.
Click to expand...

Great thank you, once I get everything set up hopefully I can help some people, but I have a feeling the way construction is going you'll be done before me!
 
  • Like
Reactions: CanCuba and Cooltiger
Is there an AI powered 1/1.8 Dahua that's the equivalent of a 1/2.8 80mm? And one that's about 170 degree wide FOV?
Large land, country side, with IR because we have super dark nights some times.
 
You must log in or register to reply here.
Top Bottom