Camera Advice
- Thread starter N49atv
- Start date
pete_c
Guru
No for IP HD; yes for analog (640X480 pixel size views).
Here decided on using fixed 2.8mm lens for outside stuff rather than the variable focal length. The granularity of the HD pixel size stuff makes up for the LD pixel sizes of yesteryear. That is me. That said I would probably utilize the variable lens (optical versus digital) if it was electronic and remote controllable maybe?
Here decided on using fixed 2.8mm lens for outside stuff rather than the variable focal length. The granularity of the HD pixel size stuff makes up for the LD pixel sizes of yesteryear. That is me. That said I would probably utilize the variable lens (optical versus digital) if it was electronic and remote controllable maybe?
tmbrown97
Senior Member
I'll try the VF one of these days for a client - but I find them less attractive than some of the other models. I have some of the 3610_FHD Dome models and I have a couple of the GXV3672_FHD models (they come in two different focal lengths now - I have both).
The original GXV3672 is WAY too narrow for anything I do - I'm glad they offer the other lens now. Coupled with their very affordable new NVR and this could be a good entry level option. However, until we see an iOS/Android app it's way behind. I hope they catch up soon.
The original GXV3672 is WAY too narrow for anything I do - I'm glad they offer the other lens now. Coupled with their very affordable new NVR and this could be a good entry level option. However, until we see an iOS/Android app it's way behind. I hope they catch up soon.
pete_c
Guru
Relating to the Grandstream GXV3672 here took one apart in an effort to change the lens. It is well constructed.
I have bench tested and taken apart most of the Grandstream IP HD cameras, DIY'd Granstream IP HD boards / cases and current Grandsteam OS cams. They are very modularly constructed. They also have SIP and VoIP built into the OS.
That is what I liked too about the older Optex combo cameras; modularity and audio built in.
Grandstream offers free NVR software called GSurf Pro for their cams which has improved much with every revision. I've been playing with it trying always to break it.
There are a few software packages that work well with Grandstream cameras. Here I do snapshots from the software or the cameras or both along with the video streaming stuff. Ideally you will want to record events rather than just do snapshots using your IP cams.
Most of the following software has Android or Iphone clients which are free.
https://www.grandstream.com/partners/vms-interoperability-partners
Primarily here though using ZoneMinder (http://zoneminder.com) (http://zoneminder.com) (http://zoneminder.com) which works well. There is available applications (free) for IPhones, Android and Wintel for ZM. Today use mostly analog sensors to drive ZM events. It does record multiple streams just fine. ZM is very flexible and works with just about any analog or IP camera today.
I have bench tested and taken apart most of the Grandstream IP HD cameras, DIY'd Granstream IP HD boards / cases and current Grandsteam OS cams. They are very modularly constructed. They also have SIP and VoIP built into the OS.
That is what I liked too about the older Optex combo cameras; modularity and audio built in.
Grandstream offers free NVR software called GSurf Pro for their cams which has improved much with every revision. I've been playing with it trying always to break it.
There are a few software packages that work well with Grandstream cameras. Here I do snapshots from the software or the cameras or both along with the video streaming stuff. Ideally you will want to record events rather than just do snapshots using your IP cams.
Most of the following software has Android or Iphone clients which are free.
https://www.grandstream.com/partners/vms-interoperability-partners
Primarily here though using ZoneMinder (http://zoneminder.com) (http://zoneminder.com) (http://zoneminder.com) which works well. There is available applications (free) for IPhones, Android and Wintel for ZM. Today use mostly analog sensors to drive ZM events. It does record multiple streams just fine. ZM is very flexible and works with just about any analog or IP camera today.
pete_c said:
According to the Grandstream literature, they have a "Vari-focal HD lens (2.8mm-12mm) with DC-IRIS and the capability to zoom or focus without opening the case". Does this mean you can control these elements from a software interface? It seems to indicate that it is software controlled, but I don't know for sure.
pete_c
Guru
I was originally going to purchase that model camera (GXV3674_FHD_VF) and didn't really liked the way it looked.
I recall posting about looking for my Aircam replacement and choosing the GXV3674 in my little blog about this stuff.
Instead purchased a Grandstream OS 3MP POE with a fixed 2.8 lens.
That said the OS does include the Vari-focal stuff (never paid attention before); so yes it would be software controlled.
Apologies for my error. There are pinouts on the camera motherboard for a vari-focal lens connection.
All of the motherboards with the Granstream OS are very similiarly constructed and work with a variety of lens type setups.
BTW next Grandstream OS camera purchase will be a 3MP dome with a 16mm lens (170 degree) for a fisheye view of the garage.
I recall posting about looking for my Aircam replacement and choosing the GXV3674 in my little blog about this stuff.
Instead purchased a Grandstream OS 3MP POE with a fixed 2.8 lens.
That said the OS does include the Vari-focal stuff (never paid attention before); so yes it would be software controlled.
Apologies for my error. There are pinouts on the camera motherboard for a vari-focal lens connection.
All of the motherboards with the Granstream OS are very similiarly constructed and work with a variety of lens type setups.
BTW next Grandstream OS camera purchase will be a 3MP dome with a 16mm lens (170 degree) for a fisheye view of the garage.
tmbrown97
Senior Member
Pete - in your dismantling of the Grandstream cameras did you find that the lens could be changed or would that likely result in breakage? For the domes I'm putting up (hopefully today if the wife allows!) I'd like to get a slightly wider view.
I'll order one of the varifocals just to test out - if it's controllable via software that could be huge for one of my regular clients.
I'll order one of the varifocals just to test out - if it's controllable via software that could be huge for one of my regular clients.
pete_c
Guru
I tried this with the very first Grandstream bullet GXV3672 camera. It looked easy when it was dismantled.
It was not as I found a replacement wider angle lens but the barrel of the lens was too long and went to the electronic shutter parts on the bottom while focusing the camera. I think I took pictures here. The thin glass shutter could break. Try it yourself but remove the upper lens barrel holder first. Its all very modular.
The dome cameras are much easier and more modular. I have bits and pieces here of the lens / IP boards while also have a bunch installed now in the Speco domes. I also have one dome purchased with a Grandstream OS. It is only 1.2MP though and I need to replace it.
I have now read that you can get a replacement lens from the same company that I used for the Ubiquiti replacement lens (I used 2.8mm lens replacements - but the edges were blurry).
I have to find the link for the lens supplier. You can just write to them with the specifics of the Grandstream lens and ask for same lens at 2.8mm or whatever. (IE: width, length type of stuff) Note that focusing the lens live in a test environment is a real PITA.
Yes I could see this being a great selling point. Here I leave mine alone and use the view to overlay variables and stuff. I play certain variables in the field of view. If I change it then I have to change the positions of my variables which is done by the pixel placement (HD IP is a lot of pixels to play with.)
The 5 MP wide screen (170 degree view fisheye) dome I am getting doesn't have IR and is an indoor camera for the garage.
Look here for the link for the lens supplier:
UK Lens Supplier
JohnDennis
Member
The one thing I notice about security camera footage, is that it only illustrates what happened and some very vague characteristics of the perpetrators. Security cameras rarely capture usable information such as license plates or facial features.
I am considering a more innovative approach.
wireless IP camera with telephoto lens mounted on my neighbors house pointed at my front door.
pinhole camera mounted at eye level at my front door to capture faces.
camera hidden in the flower bed pointed at the driveway
camera aimed at the street to record passing cars
I don't know how practical these ideas are, but I don't need any video footage of the tops of peoples heads.
I am considering a more innovative approach.
wireless IP camera with telephoto lens mounted on my neighbors house pointed at my front door.
pinhole camera mounted at eye level at my front door to capture faces.
camera hidden in the flower bed pointed at the driveway
camera aimed at the street to record passing cars
I don't know how practical these ideas are, but I don't need any video footage of the tops of peoples heads.
Your application: Practical, no. Possible, not always.
The biggest problem with the video that is often shown and touted as "footage" is usually on smaller level systems very limited applications. Many systems aren't going to be recording much over 1-2 CIF and 15FPS....it's just what is out there. I can get a heck of a lot better image and performance on a system that is running at least 4CIF and 15+FPS, but the storage (and bandwith) will be the issue. The basic facts out there are the majority of the systems out there lack proper analytics and/or appropriate storage for the system that isn't throttled to the point of only providing basic situational information.
I can put a 3-5 MP camera out there, and assuming it's not a huge area, pick up the license plate and be usable, but the issue is going to be the amount of bandwidth and storage a camera providing that level of data (assuming minimal pixel distortion at distance).
Put it this way, from doing this professionally on the enterprise level, I've got cameras installed that 99% of the public would never be able to pick up and see, but would the average install use or have a product installed in this way, not usually.
The biggest problem with the video that is often shown and touted as "footage" is usually on smaller level systems very limited applications. Many systems aren't going to be recording much over 1-2 CIF and 15FPS....it's just what is out there. I can get a heck of a lot better image and performance on a system that is running at least 4CIF and 15+FPS, but the storage (and bandwith) will be the issue. The basic facts out there are the majority of the systems out there lack proper analytics and/or appropriate storage for the system that isn't throttled to the point of only providing basic situational information.
I can put a 3-5 MP camera out there, and assuming it's not a huge area, pick up the license plate and be usable, but the issue is going to be the amount of bandwidth and storage a camera providing that level of data (assuming minimal pixel distortion at distance).
Put it this way, from doing this professionally on the enterprise level, I've got cameras installed that 99% of the public would never be able to pick up and see, but would the average install use or have a product installed in this way, not usually.
JohnDennis
Member
Is there any way to use software to capture a few stills from each vehicle based on sharpness and discard the rest? this would save a lot of space and still provide evidence.
You can't change how the video streams and the data as it presents, it needs to be present all the time otherwise there's nothing to retain or discard.
Motion based and alarm based recording. Pre/post alarm settings that increase the FPS and resolution, but you would need to pay attention to location and masking to eliminate the overall bandwidth and storage usage.
Same goes with the codecs used by the camera and system for storage. A MJPEG or MP4V will use more storage than H264.
Motion based and alarm based recording. Pre/post alarm settings that increase the FPS and resolution, but you would need to pay attention to location and masking to eliminate the overall bandwidth and storage usage.
Same goes with the codecs used by the camera and system for storage. A MJPEG or MP4V will use more storage than H264.
