Auto tracking Performance - SD4 vs SD5 vs SD8

amdpowered

Getting the hang of it
Joined
Apr 25, 2020
Messages
64
Reaction score
48
Location
USA
Does anyone have experience with the auto tracking between something like, sd49225xa-hnr vs sd5a425xa-hnr vs something in the SD8 family? I have a sd49225xa-hnr, and it seems to lose auto tracking if the object moves too fast, moves behind an object, or if the camera thinks the object has stayed still for too long. I think these are the known limitations for the SD4 line.

Is the SD5 much better with newer firmware than the SD4? How about the SD5 vs SD8?
 

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
24,868
Reaction score
48,503
Location
USA
This is as close as it gets so far:



But your issues are going to be present regardless of which PTZ as those limitations exists across all auto-tracking PTZs, including much more expensive Hikvisions being discussed here.

As I am sure you have found out, these are not plug and play. You have to dial them in to your field of views. But once you do, they are incredible. Mine hits the mark over 95% of the time. And some field of views are problematic as well.

But yes, autortrack PTZs once the tracking stops will either return to the preset it started at or with a VMS system you could tell it to go somewhere else. It is why many of us use spotter cams to help get it looking the right way if an object stops and starts again.

You can usually get it to not loose tracking so quickly if you change the target ratio a bit or turn PFA off, and I think this thing learns LOL. If the target ratio is too large, it will miss faster moving objects and lose tracking faster when an object stops.

My 49225 when I first got it would immediately lose track if someone stopped. Now it will hold at least 10 seconds if someone stops.

Do not run default/auto settings.

It is a matter of getting the brightness/contrast and target ratio settings correct.

I always knew that you shouldn't chase a bright picture - it looks nice and people migrate towards a brighter TV for example, but upon closer examination, most images need to be toned down in order to get all the details. You will be surprised how much changing a parameter like gamma could impact tracking. For example, if you have a pesky tree or something in the middle of the view during an autotrack, just by changing some image parameters you can get autotrack to pass it. Making the image a little darker at night actually helped with tracking someone across the street, which was opposite of what I thought you would think to do. So add some contrast to your image and see if it improves.

I have a yard lamp post that more times than not autotrack would get stuck on it as someone was walking and the autotrack would only go so far. Because my image has soo much contrast (bright white concrete a third, blacktop road a third, grass a third), knocking down the gamma made the lamp post not be so "trackable" lol, and along with that I turned of PFA and that gave it just enough time to retrack the person walking past the lamp post. The camera may still autotrack the lamp post when a small kid goes by, but an adult it was autotracking past the lamp post.

Ideally for an intrusion box or tripwires, you should have the initial field of view be such that the camera doesn't have to initially pan too much up/down or left/right to get the object in the center of the screen to start tracking. The closer the object is to the center of the image, the better the chance that it will track correctly.

The reason it starts looking upward or left or right is usually because the intrusion box is too big so the camera identifies the object before it is in the center of the field of view and then sometimes something else matches the "algorithm signature" of the initial object and then starts trying to track something that isn't there. Adjusting the field of view and the locations of the IVS rules to be closer to the center can fix that.

Autotracking PTZs are great, but they have limitations like everything else. Installed in a wrong location or with fields of view that do not give it a chance will be problematic.
 

amdpowered

Getting the hang of it
Joined
Apr 25, 2020
Messages
64
Reaction score
48
Location
USA
I have spotter cams that do signal the PTZ to go to another preset. It works very nicely. I'll have to play around with the gamma / contrast. I've tweaked the shutter speed settings to allow capture of fast moving plates. Currently the target ratio is set at the default 25. I might lower it to see if it'll help it not lose track.
 

tibimakai

Known around here
Joined
May 8, 2017
Messages
1,020
Reaction score
513
Location
Los Angeles
Wittaj, could you please post a picture with the target box that you use? Mine, sometimes won't even move at all, if a car is to fast, or just goes to a weird spot.
 

FlipNJ

Pulling my weight
Joined
May 3, 2016
Messages
346
Reaction score
127
Mine (49225XA) is roughly 80% hit as far as tracking. I must have tweaked it 20,000 times to get the best results. Every field of view is different so what works for one may not work for others. Like it was mentioned here, use other cameras as spotters. This does not work for me because of headlight beams lighting up areas the spotters are viewing and it immediately triggers the ptz to that area. Meanwhile, the ptz misses targets from its home position. I also have an annoying street light pole that is bright in color and if it is sunny out and later in the afternoon, the pole distracts the tracking. I also have a huge ass tree that halts the track of targets going up the sidewalk and then eventually passing behind the tree. The ptz immediately just hangs and stays on the tree for a little bit. Once in a while, if the conditions are just right, targets are tracked past the tree. The target, usually people, walk fast enough. Great all around ptz and I may grab another one for the yard. Unfortunately my ptz has a defective circuit board and the SD card cannot be recognized. I chose not to ship for service because I use Blue Iris. I do use the sd cards for local storage in case of power outages on all my cams and all the cams run on an uninterrupted power supply. They will run for about 2 hours on it. The ptz of course would not save anything because of the defective board.
 

mattp

Getting comfortable
Joined
Mar 18, 2022
Messages
339
Reaction score
967
Location
Louisiana
Mine (49225XA) is roughly 80% hit as far as tracking. I must have tweaked it 20,000 times to get the best results. Every field of view is different so what works for one may not work for others. Like it was mentioned here, use other cameras as spotters. This does not work for me because of headlight beams lighting up areas the spotters are viewing and it immediately triggers the ptz to that area. Meanwhile, the ptz misses targets from its home position. I also have an annoying street light pole that is bright in color and if it is sunny out and later in the afternoon, the pole distracts the tracking. I also have a huge ass tree that halts the track of targets going up the sidewalk and then eventually passing behind the tree. The ptz immediately just hangs and stays on the tree for a little bit. Once in a while, if the conditions are just right, targets are tracked past the tree. The target, usually people, walk fast enough. Great all around ptz and I may grab another one for the yard. Unfortunately my ptz has a defective circuit board and the SD card cannot be recognized. I chose not to ship for service because I use Blue Iris. I do use the sd cards for local storage in case of power outages on all my cams and all the cams run on an uninterrupted power supply. They will run for about 2 hours on it. The ptz of course would not save anything because of the defective board.
I'm nowhere near 80% I tweaked it and messed it up. I need to get it back to ~50%. The neighbors live oak shadow confuses the hell out of this thing...I'll take it if it loses people behind trees, but shadow from a tree?
 

mattp

Getting comfortable
Joined
Mar 18, 2022
Messages
339
Reaction score
967
Location
Louisiana
This is as close as it gets so far:



But your issues are going to be present regardless of which PTZ as those limitations exists across all auto-tracking PTZs, including much more expensive Hikvisions being discussed here.

As I am sure you have found out, these are not plug and play. You have to dial them in to your field of views. But once you do, they are incredible. Mine hits the mark over 95% of the time. And some field of views are problematic as well.

But yes, autortrack PTZs once the tracking stops will either return to the preset it started at or with a VMS system you could tell it to go somewhere else. It is why many of us use spotter cams to help get it looking the right way if an object stops and starts again.

You can usually get it to not loose tracking so quickly if you change the target ratio a bit or turn PFA off, and I think this thing learns LOL. If the target ratio is too large, it will miss faster moving objects and lose tracking faster when an object stops.

My 49225 when I first got it would immediately lose track if someone stopped. Now it will hold at least 10 seconds if someone stops.

Do not run default/auto settings.

It is a matter of getting the brightness/contrast and target ratio settings correct.

I always knew that you shouldn't chase a bright picture - it looks nice and people migrate towards a brighter TV for example, but upon closer examination, most images need to be toned down in order to get all the details. You will be surprised how much changing a parameter like gamma could impact tracking. For example, if you have a pesky tree or something in the middle of the view during an autotrack, just by changing some image parameters you can get autotrack to pass it. Making the image a little darker at night actually helped with tracking someone across the street, which was opposite of what I thought you would think to do. So add some contrast to your image and see if it improves.

I have a yard lamp post that more times than not autotrack would get stuck on it as someone was walking and the autotrack would only go so far. Because my image has soo much contrast (bright white concrete a third, blacktop road a third, grass a third), knocking down the gamma made the lamp post not be so "trackable" lol, and along with that I turned of PFA and that gave it just enough time to retrack the person walking past the lamp post. The camera may still autotrack the lamp post when a small kid goes by, but an adult it was autotracking past the lamp post.

Ideally for an intrusion box or tripwires, you should have the initial field of view be such that the camera doesn't have to initially pan too much up/down or left/right to get the object in the center of the screen to start tracking. The closer the object is to the center of the image, the better the chance that it will track correctly.

The reason it starts looking upward or left or right is usually because the intrusion box is too big so the camera identifies the object before it is in the center of the field of view and then sometimes something else matches the "algorithm signature" of the initial object and then starts trying to track something that isn't there. Adjusting the field of view and the locations of the IVS rules to be closer to the center can fix that.

Autotracking PTZs are great, but they have limitations like everything else. Installed in a wrong location or with fields of view that do not give it a chance will be problematic.
WOW, thanks @wittaj, I'll have to play with that this weekend!
 

mattp

Getting comfortable
Joined
Mar 18, 2022
Messages
339
Reaction score
967
Location
Louisiana
Unfortunately, I didn't get to this at noon like I planned, but I tweaked the settings on my 49225 a little to this:
1652568810607.png
Here's what the camera shows:
1652568866132.png
I'm worried about the tree on the right during the day, but moved the camera over to get the other tree to show for an example. Hey @wittaj, do you have any pointers on how to change the settings above to get better results? This is a first stab I'm taking at this to get better results...I guess during the day, it being washed out isn't that much of a problem, as long as you can track and identify, which is done at night in B&W. So, I could potentially make more drastic changes.

For comparison here's the same area with everything at 50%:
1652569294921.png
 

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
24,868
Reaction score
48,503
Location
USA
As long as your target/tracking ratio isn't too high of a number, it should get past those trees. Try 35 and then move up or down based on that.

That is a tough Field of View (FoV) with the shadows and the sun. You could turn WDR on, but use as low as you can, like 1 or 2, but try for no more than 10.

Conversely, do not be afraid to take the numbers below 50. Many of mine are in the 40s for the daytime. Also try to have an 8-10 point difference between brightness and contrast (with contrast higher). So something like 43 brightness and 53 contrast.
 

mattp

Getting comfortable
Joined
Mar 18, 2022
Messages
339
Reaction score
967
Location
Louisiana
Thanks @wittaj,
I was getting a little better results with what I'd done above than before. I changed the brightness to 43 and contrast to 53 (I had them flipped).

I don't remember where I got the target tracking number, probably from one of your other posts, but it's already at 35. I can tweak that a little based on results.

I don't know what I'm doing wrong, but I can't find WDR. Where's that setting? I know I've seen it but I can't find it now.

I think I have the exposure tweaked for night to make sure I get a reasonable shutter speed. But should the daytime exposure be set to manual, too? It's on Auto now:
1652634031561.png
I think having the camera adjust shutter speed throughout the day may be better than setting it though because of clouds and such, but that's just a thought.
 
Top