Managing evidence for court proceedings - system specification advice

NsteveinZ

n3wb
Oct 30, 2021
1
0
New Zealand
I need to set up a system to photograph a "stalker" and his mates, which are causing problems, typically once or twice a day. The stalker has had legal problems elsewhere, and has used guns, chainsaws and baseball bats to intimidate. Unfortunately, his home is next door... The legal strategy is to capture enough incidents to move the offending from civil to criminal offending, so as to generate jail time.

Im familiar with Dahua, so starting there. Needless to say, I need court quality images that clearly identify the offender, night or day. I need a minimum of 4 cameras to get the evidence and protect the system itself, two of which are probably going to be the evidence cameras. I don't want to spend more than 30 minutes a day processing recordings. Unfortunately there's a lot of foot traffic, so motion triggers aren't going to cut it to make the recordings manageable, there's probably 6 hours per camera of human motion, plus uncut grass etc. I can't steer the cameras during an event, so would benefit from a level of automation (e.g. zoom, human tracking). The stalker has been seen in the 20Hrs between 7am and 3am. The events happen between 5m and 15m from the cameras, but closer for an attack on the cameras. I'm expecting to buy 2.8 lenses to get a 110 FOV, any smaller and the amount of feeds to review grows....

The threat is likely to never go away, e.g. the system will be needed for many years.

The concept i'm hoping to realise is to use face and vehicle detection to index the recordings, possibly ANPR as well, to get the review time to a manageable level.

I'm heading towards:
  1. 2x IPC-HDW5541H-ASE-PV (5mp), or IPC-HDW5541H-ASE(8mp) monitoring other boundaries to capture vandalisation of the main cameras from outside the FOV?
  2. 2x IPC-HDBW71242H-Z (12mp evidence cameras)
  3. NVR ...
My questions are:
  1. Is this adequate or should I up-spec further?
  2. Does a WizMind NVR add value when you have WizMind cameras, or is it essential?
  3. What do you get for face detection by camera? Is it just a metadata flag added to the frame, that the NVR can index?
  4. Can any system identify a person who walks with their face obscured, etc.?
  5. Does face detection by camera identify the person, or is identification done by the NVR as real-time or post processing.
  6. Can face detection drive a motorised zoom?
  7. Can face detection pivot the camera and later return it to the normal position?
  8. Is 5mp adequate for closework?
  9. Assuming the 4 cameras feed in using H.265 ,do I need to worry about a system bandwidth of only 80 Mbps?
  10. On the assumption I might need a 5th POE, should I go 2*4POE NVR or 1*8POE NVR?
  11. I have one approach I haven't got covered (over the flat roof) of the single storey building that accommodates the cameras. May need two additional ports in the future, to get 360 degree protection of the host building.
  12. Can I filter the recording by face and/or vehicle plate number?
  13. Can I filter recordings to classify new faces and vehicle plates?
 
Welcome!

Here are a few guidelines and considerations as you piece something together.

It is simple LOL do not chase MP - do not buy a 4MP camera that is anything smaller than a 1/1.8" sensor. Do not buy a 2MP camera that is anything smaller than a 1/2.8" sensor. Do not buy a 4K (8MP) camera on anything smaller than a 1/1.2" sensor. Unfortunately, most 4k cams are on the same sensor as a 2MP and thus the 2MP will kick its butt all night long as the 4k will need 4 times the light than the 2MP... 4k will do very poor at night unless you have stadium quality lighting (well a lot of lighting LOL). Starlight, ColorVu, Full Color, etc. are simply marketing terms, so don't be sold on those names.

To identify someone with the wide-angle 2.8mm lens that most people opt for, someone would have to be within 13 feet of the camera, but realistically within 10 feet after you dial it in to your settings.

1635651807366.png

My neighbor was bragging to me how he only needed his four 2.8mm fixed lens 4k cams to see his entire property and the street and his whole backyard. His car was sitting in the driveway practically touching the garage door and his video quality was useless to ID the perp not even 10 feet away. Meanwhile my 2MP varifocal optically zoomed in to the public sidewalk provided the money shot to the police to get my neighbors all their stuff back. Nobody else had video that could provide anything useful, other than what time this motion blur ghost was at their car.

Here are my general distance recommendations, but switch out the Dahua 5442 series camera to the equivalent 2MP on the 1/2.8" sensor or equivalent Hikvision works as well.
  • 5442 fixed lens 2.8mm - anything within 10 feet of camera OR as an overview camera
  • 5442 ZE - varifocal - distances up to 40-50 feet (personally I wouldn't go past the 30 foot range but I like things closer)
  • 5442 Z4E - anything up to 80-100 feet (personally I wouldn't go past 60 feet but I like things closer)
  • 5241-Z12E - anything from 80 feet to almost 200 feet (personally I wouldn't go past 150 feet because I like things closer)
  • 5241-Z12E - for a license plate cam that you would angle up the street to get plates up to about 175 feet away, or up to 220 with additional IR.
  • 49225 PTZ - great auto-track PTZ and in conjunction with an NVR or Blue Iris and the cameras above that you can use as spotter cams to point the PTZ to the correct location to compliment the fixed cams.
You need to get the correct camera for the area trying to be covered. A wide angle 2.8mm to IDENTIFY someone 40 feet away is the wrong camera regardless of how good the camera is. A 2.8mm camera to IDENTIFY someone within 10 feet is a good choice OR it is an overview camera to see something happened but not be able to identify who.

One camera cannot be the be all, see all. Each one is selected for covering a specific area. Most of us here have different brands and types, from fixed cams, to varifocals, to PTZs, each one selected for it's primary purpose and to utilize the strength of that particular camera.

So you will need to identify the distance the camera would be from the activities you want to IDENTIFY on and purchase the correct camera for that distance as an optical zoom.

Main keys are you can't locate the camera too high (not on the 2nd story or above 7 feet high unless it is for overview and not Identification purposes) or chase MP and you need to get the correct camera for the area trying to be covered. A 2.8mm to IDENTIFY someone 40 feet away is the wrong camera regardless of how good the camera is. A 2.8mm camera to IDENTIFY someone within 10 feet is a good choice OR it is an overview camera to see something happened but not be able to identify who. Also, do not chase marketing phrases like ColorVu and Starlight and Full Color and the like - all cameras need light - simple physics...

If you want to see things far away, you need optical zoom, digital zoom only works in the movies and TV...And the optical zoom is done real time - for a varifocal it is a set it and forget it. You cannot go to recorded video and optically zoom in later, at that point it is digital zoom, and the sensors on these cameras are so small which is why digital zoom doesn't work very well after the fact.

Regarding a camera for plates (LPR) - keep in mind that this is a camera dedicated to plates and not an overview camera also. It is as much an art as it is a science. If you want to capture plates, you need a camera just for that.

During daytime it may act as an overview, but at night the shutter tends to need to be faster to capture the plate, and the faster the shutter, the darker the image.

Most of us run B/W to take advantage of the reflective properties of a plate which is why we can run a faster shutter to get the plates. Once you decide to keep it in color, the shutter speed slows way down and might not be fast enough to actually capture a plate.

You may have enough light and the car moving slower to capture, but you may also need to decide what is more important - the overview or the plate. Most of us run two cameras.

At night, we have to run a very fast shutter speed (1/2,000) and in B/W with IR and the image will be black. All you will see are head/tail lights and the plate. Some people can get away with color if they have enough street lights, but most of us cannot. Here is a representative sample of plates I get at night of vehicles traveling about 45MPH at 175 feet from my 2MP camera (that is all that is needed for plates):

1635651834838.png

Keep in mind that most of us have found in home settings that facial recognition is more gimmicky and novelty than anything else. If you have to put in 5 or 10 or 15 or 35 pictures of yourself in system for it to recognize it is you...then you shouldn't expect much.... My success rate was under 5% so I moved on to other things LOL.

It can work in certain situations like a business that requires everyone to stop in front of the camera and the camera is at head height. Outside of that, the percentage of being accurate is probably not going to be super high. Especially someone coming to a front door.

Unless you spend the big bucks that casinos and airports have LOL.

Heck even in ideal situations like a business with the camera at ideal height and optimal lighting it fails....



Answers to your questions are:
  1. Is this adequate or should I up-spec further? You will want more cameras. No question...
  2. Does a WizMind NVR add value when you have WizMind cameras, or is it essential? If you want full functionality of the camera to work with the NVR, then yes.
  3. What do you get for face detection by camera? Is it just a metadata flag added to the frame, that the NVR can index? Yes, but see above.
  4. Can any system identify a person who walks with their face obscured, etc.? No, only in the movies.
  5. Does face detection by camera identify the person, or is identification done by the NVR as real-time or post processing. Somewhat real time, but see above.
  6. Can face detection drive a motorised zoom? With an auto-track PTZ, it can track human or vehicle.
  7. Can face detection pivot the camera and later return it to the normal position? Autotrack PTZ will return to home/normal/cycle after event is over.
  8. Is 5mp adequate for closework? See above. If the 5MP is on same sensor that a 2MP is on, the 2MP will be better.
  9. Assuming the 4 cameras feed in using H.265 ,do I need to worry about a system bandwidth of only 80 Mbps? You will get more cameras, so no 80Mbps is too low. My neighbor has 4K cameras and has to run at a low bitrate due to limit of NVR bitrate.
  10. On the assumption I might need a 5th POE, should I go 2*4POE NVR or 1*8POE NVR? Go with minimum 8, but maybe consider 16.
  11. I have one approach I haven't got covered (over the flat roof) of the single storey building that accommodates the cameras. May need two additional ports in the future, to get 360 degree protection of the host building. Plan now for a larger NVR than you need
  12. Can I filter the recording by face and/or vehicle plate number? Yes, but recognize the limitations. See above.
  13. Can I filter recordings to classify new faces and vehicle plates? Yes, but recognize the limitations. See above.
 
need to see a house property layout and where you plan on mounting the cameras. A google satellite picture would help.

Mount the cameras low, 2 meter up, no more than 3 meters up.
LPR cameras are specific to License plates, then do nothing else only license plates. The best LPR setup is to use two cameras, an over view and then the license plate camera.
Each cameras MUST be covered by another camera. If the the camera is destroyed, you must have video of it being destroyed.
The time stamps on the cameras must be very accurate, within 1 or 2 seconds.
More cameras are better. As an example I have 3 cameras on the outside of the front door, and one inside point at the front door from inside.
 
Advice given above is dead on. Do NOT chase megapixels!! More cameras, and specific cameras for specific applications is the way to go. Do not underestimate the number of cameras required for proper coverage. As @SouthernYankee said, put up a satellite picture, or at least map with buildings/obstructions clearly located. If you want license plates, do not expect a camera to capture them and other video at the same time, regardless of camera resolution. I have two LPR cameras one a 2 MP at 130 feet, and one a 4 MP at 40 feet. The 2 MP is zoomed in to get license plates, works day and night. The 4 MP is a 6 mm lens, but just barely gets them during the day, and none at night.

My suggestion for you right NOW:

1. Buy a single 5442 varifocal.
2. Buy a used eBay 6th generation or better i5/i7, Windows 10 or preferably, 10 Pro - should be $300-$500. See the Wiki for hardware recommendations
3. Add or get 16 GB ram, a 240 GB SSD to boot, and a WD purple HD probably 4-6 TB to start.
4. Buy BI, install it and get the whole damned thing up and running to understand cameras, views, etc.
5. Learn from the above information, and figure out additional camera requirements.
 
2.8mm means the perp needs to be quite close -- you can't have your cake and eat it to there (face capture at a distance and wide FOV...)

Lighting: what's your lighting situation, spend some of the big 12MP PTZ $$ on some lighting to help out with the money you'll save by likely not getting that camera. For evidence, ideally you'll want color nightitme shots and lighting will help that immensely.
 
Last edited: