Camera Slow Shutter / Ghosting Tested

By: John Honovich, Published on Jan 17, 2014

Wrong. Your first impression of an IP camera's performance in low light is frequently wrong.

The reason is simple: slow shutter.

Many manufacturers default their cameras to go to a shutter speed that is too slow to capture moving objects without creating blur/ghosting effects.

They do this because they correctly assume that most users will 'test' by pointing their cameras at a wall and judge performance by how bright the image is. Only later when it is put in production and cars or people pass by, do they notice the issue.

Manufacturer Slow Shutter Ranges

Manufacturer's default slow shutter setting typically ranges from 1/4s to 1/30s and are often given fancy marketing terms to obscure what they are doing. This is a huge deal as the cameras on the slower end (i.e., 1/4s) are taking in 750% more light than the ones of the higher end.

Showing It In Action

We tested 6 manufacturers using a comparative approach:

  • On the far left, is a stationary person with slow shutter on.
  • In the center left, is a moving person with slow shutter on.
  • In the center right, is a stationary person with slow shutter OFF. 
  • In the far right, is a moving person with slow shutter OFF

Here's an Arecont camera, showing the tradeoff:

*****. **** ***** ********** of ** ** ******'* performance ** *** ***** is ********** *****.

*** ****** ** ******: slow *******.

**** ************* ******* ***** cameras ** ** ** a ******* ***** **** is *** **** ** capture ****** ******* ******* creating ****/******** *******.

**** ** **** ******* they ********* ****** **** most ***** **** '****' by ******** ***** ******* at * **** *** judge *********** ** *** bright *** ***** **. Only ***** **** ** is *** ** ********** and **** ** ****** pass **, ** **** ****** the *****.

Manufacturer **** ******* ******

************'* ******* **** ******* setting ********* ****** **** 1/4s ** */*** *** are ***** ***** ***** marketing ***** ** ******* what **** *** *****. This ** * **** deal ** *** ******* on *** ****** *** (i.e., */**) *** ****** in ***% **** ***** than *** **** ** the ****** ***.

Showing ** ** ******

** ****** * ************* using * *********** ********:

  • ** *** *** ****, is * ********** ****** with **** ******* **.
  • ** *** ****** ****, is * ****** ****** with **** ******* **.
  • ** *** ****** *****, is * ********** ****** with **** ******* ***. 
  • ** *** *** *****, is * ****** ****** with **** ******* ***

****'* ** ******* ******, showing *** ********:

 

[***************]

**** ***

**** **** ******* ******* to */** **** *******. In *** *** ********** below, ** *** *** our ******* ** * blur ***** **** *****. At */***, *** ***** becomes ********* ** **** *** *** **** ******* is ********* ** ******.

*******

******* ******** ** ** exposure **** ****** ***** "balanced", **** ** ********** of **** ******* ***** it ******** ** (*/**). Worse, ********** ***** **** the ******* *******, *** the *** **** ** a **** *** ****** when ***** "*********" ****.

** */** ******* *****, images *** ********** ******** ** *** AV3116, ****** ** ******* of *** ****** ******* are ********* *** ** motion ****. ******** ******* ***** ** */***, the ***** **** ********, but **** ** **********, providing ***** ****** *******.

******** ** *** *** version, ***** *** ******* is **** *****:

**** *****

*** **** ***** *** ***** **** *** ** ******** setting ****** "*******", ***** equates ** */* ****** shutter. ** **** *******, our ******* ** * severely ******* *** *** barely ** ****. ******** ******** ** */***, *** image ******* *** ***** increases *******, *** **** details ** *** ******* such ** ******** *** hair ***** *** ** made ***. **** **** the ***** *** ***** defaults ** */*** ******* speed.

***** ***** ****

*****'* ****** ***** *****, *** ****, ******** ** */** shutter *****, ***** *** refer ** ** "*****-***** exposure" ** ********. ** this *****, *** ******* is * ******** ****, almost ************** ** *****. However, ************* ** * 1/30s ******* ***** ******** a **** ****** *****, with *** ******* ********* to ****** *** ** details *******.

*******

** *******, *******'* ******* *** ******* *** */** ******* *****. In *** ***** ***-****, this ******* ** *********** ghosting, **** ******* ***** than *****/**** ******** ******** (no **** ** ******** style ***********).

*** ***-**** **** ******** to */** ******* ******* speed. *******, ** ***** that ** *** **** scene ** ***** ******* in **** ****, *** to *** ****** *** light ***********, ** *** not **** *** ******* nearly ** ****. **** blurring ** **** ***** default ********, **** ** the **** *****. ** the *****, *** *******'* face ** ********** *******.

Honorable *******

** *** *****, ** verified * ****** ** cameras ** *** ******* to **** *******, ********** accurate *** ** *** box *** ***** ***********, including (*** *** ** exhaustive ****):

  • **** *****, *****, *****
  • ***** ***-****, ***-****, ***-****
  • ********-*******, ***-*******
  • ********* **-*********, **-********, **-********
  • ********* **-*****
  • **** ***-*****, ***-******, ***-*****
  • *************, *******

Comments (20)

John,

I've found that many in my industry don't understand how slow shutter works. I'm always amused/dismayed to see cameras deployed where higher frame rates are required but slow shutter settings are not checked. On many VMS's, slow shutter does not appear as a change in the frame rate display so users assume they're seeing higher frame rates than they are actually getting.

We've had debates with people about vendors who use slow shutter but show full frame rates. It's obviously cheating. The frame rate cannot be faster than the shutter. If my shutter speed is 1/10s, that means I can only get 10 unique 'shots' per second. The only way I can deliver 30fps in that scenario is to copy each frame 3 times, which is pointless.

Very important thing. My feeling is, like Carl already mentioned, a lot of installations are done with defaullt on shutter, priotirizing fps. Live images and recordings appear to be good, job done. But just as long as no one wants to get an image or snapshot from a moving object.

In most of our projects I ask installer to do exactly this: Try to create snapshots from moving objects to find out quality of current settings. Takes more time but it's worth doing it. Keep in mind: Most interesting objects you wanna see with your cam might move ;)

Harald, good feedback. It's funny that the VMS vendor gets blamed for this sometimes. Though, I guess, the VMS vendor gets blamed for a lot!

It is a shame that so many 'professionals' have no idea about shutter speed / exposure and the impact it is.

Btw, for those looking for background references, see:

FYI all Avigilon cameras come defaulted at 1/30

Is 1/30 shutter speed deemed as the "industry standard"?

We have a project where the cameras are installed in areas with natural lighting during the day and florescent during the night based on timer from 1900 to 0700hrs. During cloudy days or early mornings, lux level is about 11 - 16 lux, but the image is too dark for identification. Shutter speed was set at 1/100.

Image brightness will only be acceptable a 1/25, but there will be motion blur when a snap shot is taken. This affects manual licence plate and face recognition.

IR LED is not an option and image must be in colour for identification purpose.

Any advise?

Good question!

The industry 'standard' default is generally to make the shutter speed match the max frame rate. To the extent that most cameras max frame rate is 30fps, that would mean 1/30s shutter. Of course, PAL and countries where PAL is a legacy, sometimes see 25fps max frame rate and accordingly, a slowest shutter of 1/25s. By contrast, for cameras supporting 60fps, this requires 1/60s shutter to deliver true and full 60fps.

If you have 10+ lux light during this period and the video is too dark with 1/100s, this indicates a poor low light camera. 10+ lux is a lot of light, relatively speaking, even at 1/100s.

What is the camera model you are using? Would you consider switching it? That or adding in a white visible light illuminator...

It's a Hikvision DS-2CD755F with min illumination spec at 0.05 lux.

We can't switch the camera as it's already in an implementation stage.

We are trying to avoid the last resort of adding the illuminators as the sites are many and it's lots of additional cost.

Ri Na- the 755 camera will deafult to 1/30 shutter and does not support slow shutter. We would need to know what other setting values are and to see snap shots of the scene you are viewing. Which specific model are you using? Please send your contact information and any pertenent settings info to me at bob.germain@hikvisionusa.com I will take a look myself and also hook you up with the tech support as nesessary.

It is simple to measure the shutter speed of any camera using any analogue oscilloscope. It is possible even to build a dependence between the shutter speed and illumination level using luxmeter and a lamp in addition.
Further, knowing the sputter speed and velocities of objects on the scene it is possible to simulate their blurring in the frame.

I live in a very high crime area with daily robberies.    What cameras or strategies would you recommend that minimize motion blur (fast shutter speed) in night time conditions?  I will probably need to install somewhere between 6 and 10 cameras.  I also want a few cameras that will take video far away (200-600 feet away ) if possible

NOTICE: This comment has been moved to its own discussion: A Very High Crime Area With Daily Robberies, What Cameras Or Strategies Would You Recommend In Night Time Conditions?

That's going to depend on your budget.  If I recall correctly, the Sony full frame 35 mm camera has the best low light performance among models tested here.  The Axis Q1359 uses an APS-C sensor and is also very good in low light with a very fast lens.  Those are very expensive options.

Next up would be some of the large sensor PTZs for that long range, also expensive.  There are tests of those here as well.  Unfortunately I can't offer first hand advice for any of the above.  For close range the best low light performer I have is a Hikvision DS-2CD5546G0-IZHS, but its focal length won't come close to allowing 200 foot use, let alone 600 feet.

I will say that in my tests to keep normal walking pace almost free of blur, you need a shutter speed of 1/120s or faster.  I have not seen any tests here where that fast of a shutter speed is used in low light.  (I could have missed it.)  I believe they have a standard of 1/30s at night.  That will work if your subject slows down.  1/60s is of course better, and obviously faster shutter speeds require more light and/or better low light performance in the camera.  Subjects walking straight at the camera may also appear less blurred even at 1/30s or 1/60s.  

At the range you mention, you will probably need some other light sources to do a good job.  Check out this test to see how a couple of the long range PTZs do:

https://ipvm.com/reports/axis-vs-hikvision-ir-ptz

For budget, I'd like to keep it well under $5,000 if possible.  I will do the install work myself.  Here's a link to the camera layout I am planning to set up. 

https://ipvm.com/calculator/2q8SOA

There are streetlights, and the home has numerous lights along the perimeter, so there's some light actually.  Some areas, like the alleys, are darker than others.  I am willing to replace the current bulbs in the perimeter lights to higher lumen bulbs if it will help.  Ideally, I'd like the setup to be able to capture people hustling or running as well, since that is what happens from time to time with our robberies.  

If you want to catch faster motion, I definitely suggest a few of the DS-2CD5546G0-IZHS for closer range.  It's a 2.8 to 12mm zoom so you have some options on range, just not out hundreds of feet.  $509 as of now at B&H.  Your next step down cost wise might be the DS-2CD2H45FWD-IZS.  About half the price at B&H, yet still with the 2.8 to 12mm zoom.  If you have external lights this might be good enough.  This one is a turret but there is probably a dome with similar features or exactly matching features.  There's a bullet style as well, but it is priced higher, at least on B&H.

For those long range views down the streets, check out the 8 to 32 mm zooms like the DS-2CD4A25FWD-IZH8.  There are others that are specific for license plates, like the DS-2CD4A26FWD-IZHS8/P bullet camera.  Don't expect to get much in the way of identifying details at 500 or 600 feet without bright light on the subject.

Chris, 

 

I pulled two of your recommended cameras up for comparison.  

https://ipvm.com/compare?utf8=%E2%9C%93&cameras%5B0%5D=7543&cameras%5B1%5D=7524

At half the price I am tempted to save the money.  Between these two what is the major reason for the price drop and what will I be missing most from the cheaper model in regard to image clarity of people who are running in the daytime and at 10 lux evening time.   Regarding VMS can you tell me in the most generic sense what the issue is with VMS and the 4 cameras you recommended?  

Thanks for your help.

The more expensive cam is better in low light and is going have more "Smart Events" options. I have no idea how either cam will perform with people running in the current night time lighting at your location.

To catch identifying details of people running, you need a faster shutter speed.  Much faster than 1/30s or 1/60s.  Think somewhere between 1/200 and 1/500. I have not seen any comparison tests showing people running at night, and I've only done tests walking around in view of my cameras, not running.  Views of my kids running past the cameras in the daytime at 1/120s are pretty much a blur.

For a faster shutter speed you need good light and/or a camera with very good low light capability.  Adding some bright flood lights plus mid range cameras might be cheaper than more expensive cameras alone, and it has the added benefit of deterrence and allowing people in the area to see what's going on. 

Most Hiks work with several different VMSs, but I have never used a VMS.  The question is how do you plan to store and manage the video from the cameras?

FYI, positive votes on comments helps with IPVM subscription fees.

Here's an example of what you're dealing with.  I sent one of my kids out to simulate running.  He's on a scooter going what would be a slow running pace for an adult, it's completely overcast about an hour before sunset, and he's 20 to 25 feet from the camera and headed straight at it.  Shutter speed is 1/250s, and there is only slight blur. 

This is an 8 MP OEM Hikvision that claims low light performance that is the same as the lower cost one I mentioned before.  Since the sensor is the same size, I would not be surprised to see that the 4 MP version is actually better in low light, but there's enough light in this shot that the higher resolution is useful.

Overcast at 1/250s

Make sure to also get recommendations from people that use brands other than Hikvision. You many find that a VMS is critical, or some other factor, and that's not something I can offer advice on.

This subject gives me a headache.

 

Not because it is very complicated. I learned most of this years ago, when learning how to use my big boy Canon D50 DSLR and shooting fastpitch softball. Try to capture a softball in mid air and see writing on the ball, or see the ball mush when the bat hits it...

 

But explaining this to customers. How in the world do you explain this to clients? Do you just present a few snapshots, and ask, which one is acceptable for night time performance?

 

 

NOTICE: This comment has been moved to its own discussion: How Do You Explain Slow Shutter / Ghosting To Clients?

Once attended a training course given by a prominent VMS company and had to argue with the instructor and some of my classmates about the importance of understanding the fundamentals of light capture in basic photography.  Most thought adjusting the frame rate would improve the capture.  A few had no idea what we were talking about.  Sad! ;-)

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Related Reports on Low Light

Axis "Best Of The Best" 4K Camera Tested (Q1798-LE) on Nov 21, 2019
Axis has released their "best of the best" Q1798-LE bullet camera, touting "4K without compromise" with a large Micro 4/3" image sensor, custom...
Hikvision Dual Lens Face Recognition Camera Tested on Nov 19, 2019
Hikvision's Dual Lens Facial Recognition camera, claims that it "adopts advanced deep learning algorithm and powerful GPU to realize instant face...
Color Low Light Mega Camera Shootout - Avigilon, Axis, Bosch, Dahua, Hanwha, Hikvision, Panasonic, Speco, Sony, Vivotek on Nov 12, 2019
This is the biggest color low light shootout ever, testing 20+ super low light models from 10 manufacturers: Increasingly, each manufacturer...
Biggest Low Light Problems 2019 on Nov 08, 2019
Over 150 integrators responded to our survey question: "What are the biggest problems you face getting good low-light images?" Inside, we share...
Avigilon Appearance Search Tested on Oct 30, 2019
Avigilon Appearance Search claims that it "sorts through hours of video with ease, to quickly locate a specific person or vehicle of interest...
Illustra Pro Gen3 4K Camera Tested on Oct 23, 2019
Johnson Controls has released the latest generation in their Illustra Pro line, the Pro Mini-Dome Gen 3, a non-OEM model claiming to "optimize...
Integrated IR Camera Usage Statistics 2019 on Oct 21, 2019
Virtually every IP camera now comes with integrated IR but how many actually make use of IR or choose 'super' low light cameras without IR? In...
Pelco Sarix Pro3 Camera Tested on Oct 16, 2019
Pelco has released their Sarix Professional Series 3 cameras, claiming "more security detail in challenging scenes with excellent low light and...
Avigilon H5A Analytic Cameras Tested on Oct 07, 2019
Avigilon has released its H5A analytic cameras, claiming to "detect more objects with greater accuracy even in crowded scenes." We tested the...
Wide Dynamic Range (WDR) Guide on Oct 01, 2019
Understanding wide dynamic range (WDR) is critical to capturing high quality images in demanding conditions. However, with no real standards, any...

Most Recent Industry Reports

Axis "Best Of The Best" 4K Camera Tested (Q1798-LE) on Nov 21, 2019
Axis has released their "best of the best" Q1798-LE bullet camera, touting "4K without compromise" with a large Micro 4/3" image sensor, custom...
Glass Doors and Access Control Tutorial on Nov 21, 2019
One of the biggest access challenges are locking and securing glass doors. Unlike wood or steel doors that can be modified to work with...
Avigilon H4 Intercom Tested on Nov 20, 2019
Avigilon is well-known for video surveillance and access, but how well does the company's intercom work? We purchased and tested Avigilon's H4...
The Cowardly, Greedy "Leaders" of Video Surveillance - SIA on Nov 19, 2019
The video surveillance industry suffers from cowardly, greedy 'leaders' focused on maximizing easy money while undermining public trust. The...
Hikvision Dual Lens Face Recognition Camera Tested on Nov 19, 2019
Hikvision's Dual Lens Facial Recognition camera, claims that it "adopts advanced deep learning algorithm and powerful GPU to realize instant face...
Top Manufacturers Gaining and Losing 2019 on Nov 18, 2019
2019 has been an explosive year for video surveillance, with the world's two largest manufacturers, Dahua and Hikvision, being sanctioned for human...
Hidden Camera Detectors Tested on Nov 18, 2019
Hidden cameras are a growing problem as cameras become smaller, cheaper and easier to access. However, some companies claim to be able to detect...
Wyze Fires Back at JCI - Your Patents Are Invalid, Pay All Of Our Costs on Nov 18, 2019
Goliath JCI targeted startup Wyze this summer alleging the fast-growing consumer startup was violating a slew of JCI's patents. Now, Wyze has...
ADT Stock Surges - "Leading The Commercial Space" on Nov 15, 2019
Don't call it comeback... but maybe call it a commercial provider. ADT, whose stock dropped by as much as 2/3rds since IPOing in 2018, has now...