Strange noise / pixelation at bottom of camera image

ermac

Getting the hang of it
Jun 21, 2018
105
45
London and South Africa
Hi, I have a 2mp Dahua PTZ, that has been brilliant since 2019.

However, recently, the past 2 or so months I have noticed a strange constant pixelation / noise at the bottom of the screen. It is there constantly. Whether there is movement in the scene or a still scene, change of light, shadows etc. And it remains at the entire length of the bottom of the screen, same place on zoom in and out, panning or tilting and remains the same height.

I have the bit rate set to the maximum the camera can handle.
It was never there before, only from the past 2 or so months.

I have attached a cropped MP4 video (so not the entire field of view), it is short (3 seconds) but shows what I am referring to. Please look at the bottom of the video image.

Could this be due to damage on the ethernet cable?
 

Attachments

What are you camera settings as in below image?

FlagCam_video_config.jpg
 
Consider changing H.265 to H.264 and set iframe interval to same as frame rate (I use 15 FPS with 15 frame interval).
 
I haven't changed the settings since the camera was setup, and it always gave me a great image.

So do you rule out cable damage? And think it is more setting related?
 
So do you rule out cable damage? And think it is more setting related?
No, I'd never rule that out. It's just that usually that's not as easy and quick to eliminate as are cam settings.
But since you said it's worked great up until now I'd try some thing else...like the cable.

I have a 100 ft. Ethernet patch cable just for bypassing temporarily in such instances. And while up there you can check the RJ45 connection at the pigtail for moisture and corrosion.
 
The cable I run to the PTZ is a Ubiquiti TOUGHCable but there might be a kink on a bend or something like that which I can't rule out. Issue running a test cable though is... the PTZ is on a pole 3.5m (11.5 feet) high up :lol: so bit of a challenge, but might need to be done.
 
Why would a damage in the transmission cable only affect the last few lines of the picture? POE failing?
the PTZ is on a pole 3.5m (11.5 feet) high up
A damage in the wiring to the sensor, OK I can see how that would work. The camera is likely exposed to direct sunlight much of the day. Successive heating and cooling can damage soldering joints. Moisture could also have attacked the soldering on an end. Or the sensor, or the ram. But if this is damage to data connections somewhere, it has to be where the data is still regionally separated.
my 2 cents
 
Why would a damage in the transmission cable only affect the last few lines of the picture? POE failing?

A damage in the wiring to the sensor, OK I can see how that would work. The camera is likely exposed to direct sunlight much of the day. Successive heating and cooling can damage soldering joints. Moisture could also have attacked the soldering on an end. Or the sensor, or the ram. But if this is damage to data connections somewhere, it has to be where the data is still regionally separated.
my 2 cents
That all could be the case but how would you proceed troubleshooting?
Pull down the cam, open it up and inspect/test the items/components you mentioned?
Or do something relatively simple first before proceeding to the more difficult?
Personally, I start out with the tack hammer before I pick up the sledge hammer.....something I learned from close to 50 years field and shop electronics/electrical troubleshooting and repair.....my 2 cents. :cool:
 
Last edited:
That all could be the case but how would proceed troubleshooting?
Pull down the cam, open it up and inspect/test the items/components you mentioned?
Or do something relatively simple first before proceeding to the more difficult?
Personally, I start out with the tack hammer before I pick up the sledge hammer.....something I learned from close to 50 years field and shop electronics/electrical troubleshooting and repair.....my 2 cents. :cool:
Well, if it is the soldering, the camera might be sensitive to physical shock. So I'd knock it, to see if the effect changes during shock. (Don't use the sledge hammer though.) Now this might be something one can do with a rod, fork, an athletic child... before going up oneself. (However, the only result of this would be that one has to go up oneself, no matter what the result)

If that comes up negative, there isn't much one can do without going up. You could change the PSU unit down on earth. But in order to change the cable, you are already going up. You already have to open the plug. Me personally, I would loosen the screws too and take it to my desk. And do the rest in a secure altitude.
 
  • Like
Reactions: ermac
If accessing the camera is difficult, do a factory reset. Sometimes things go wonky and a factory reset clears out the bugs. And then setup from scratch - do not import config file.
 
Why would a damage in the transmission cable only affect the last few lines of the picture? POE failing?

A damage in the wiring to the sensor, OK I can see how that would work. The camera is likely exposed to direct sunlight much of the day. Successive heating and cooling can damage soldering joints. Moisture could also have attacked the soldering on an end. Or the sensor, or the ram. But if this is damage to data connections somewhere, it has to be where the data is still regionally separated.
my 2 cents

I changed the POE+ injector for a new one. So it is not a PSU issue.
 
i have closely analyzed the video and what is appearing is artifacts of compression, from the very thin horizontal black strip at the bottom.
is there an option to enable "overscan" on the camera or dvr ?
that should eliminate the artifacts.
That setting is not on the NVR nor on the camera's settings.
 
If accessing the camera is difficult, do a factory reset. Sometimes things go wonky and a factory reset clears out the bugs. And then setup from scratch - do not import config file.
I can try this, but that means I am going to lose all my presets, I have over 40 setup :facepalm:
So no way damaged network cable would cause this?
 
Well, what about switching to h264 and giving the camera the highest possible bandwidth with that codec.
Or define a ROI in that part of the picture and see if the effect changes significantly.
It might help to support or refute the artifact theory.
 
  • Like
Reactions: ermac
Well, what about switching to h264 and giving the camera the highest possible bandwidth with that codec.
Or define a ROI in that part of the picture and see if the effect changes significantly.
It might help to support or refute the artifact theory.
I'll set to h264 and set it to 8192kbs and see...