Testing Axis' 1/2" Q1635 Camera

By Ethan Ace, Published Nov 23, 2015, 12:00am EST

Low light performance continues to improve, first driven by advances in image processing and now increasing number of 1/2" imagers in 1080p HD cameras.

IPVM has recently tested new super low light, super low cost offerings from Hikvision and Dahua, and in this report we see how Axis' 1/2" Lightfinder model, the Q1635, stacks up.

We tested it against other large imager super low light models from Dahua, Hikvision, and Samsung, and past 1/3" low light leaders from Axis and Samsung in multiple scenes, including outdoor low light scenes including a dark field area (left) and a moderately lit parking lot (left):

*** ***** ******** ******* ******* to ***** */*" ***** low ***** ****** ** night with significant ********** ** ******* in ***** ****** (*** ** *********), ***** ************* **** bandwidth **** ***** ***** low ***** ******. 

*******, ** ******** ****** ****** than ***** */*" ****** ******, which *** **** ** less ******** ** **** low ***** (***** *.* lux), ***** ******* *** not ********, **** *********. Those ******* *** **** brightest ******* ** ***** scenes *** ****** ****** served ** ********* ** Dahua */*" ******.

*** ***** ******* ** stay ** *** **** line *** ********* *** absolute **** *** ***** (non-IR) ***********, *** ***** offers ****** ******* **** any ***** **** ***** we **** ******.

*************, *** ********* **** be ******** **** ***** low ***** ***********, ****** bandwidth *** ************* ****** ******* of *** **** ** Dahua, ********* *** *******.

Notable ***** ********

*** ********* *********** **** produced *** ******* ********* ******, though ********** ********** ******** in *** ***** **** using *** **.** ********* ******* lens, ******** ******* ** the ****** ** ***** cameras.

*** */*" ***** ********* significantly better **** *** */*" Lightfinder *****, *** *****, in *** ***** ****** aside **** **** **** (<0.01lx) ***** **** ******** only * ***** *****.

*** *****, *** ** Zipstream ********, *** ** far *** ****** ******** of *** */*" ***** low ***** ******, ***** 1 **/* ** ~*.** lux. ********* *** ***** consumed *-** *** *********.

*******

*** **** ***** ***** for~$*,*** *** ******, ********* * ***** F1.5 ****.

**** ******* ** ************* higher **** **** */*" ******, such ** *****, *********, and *******, **** ** line **** ****** **** as *** ***-*****, ****'* 1/2" **********.

Physical ********

**** ***** ******* *** physical ************ ** *** Q1635. **** **** ********/******** features *** ****** ********* to *** *****, **** the ***** *********** ***** the ****** ****** *** corresponding **** (**.* *-***** with ******* ****). **** lens ** ****** ** large ** *********'* **.** DarkEye ****, **** ****** **** standard ******.

*** ***** *** ***** both ******* ******** ********** features, **** ** ***** stabilization, ***** *********, ******** assistance, ***.,******** ** *** **** of *** *****.

*** ***** ***** ***** the ******** **** ** the *****'* ***** ****** (1/2") ******** ** *** Q1615, ********** *******.

Hikvision **.** **** *******

******* *** ****** ** the *********** ***** **** there ** ********** ******** of *** ***** *** object ***** ** *** Hikvision *********** ****** **** the **.** ****. ** tested **** **** ** the **** ***** ** see ** **** ******** was *******, *** **** versa, ******* *** **** lens ** *** ********* 6026 ** *** ** images **** *******.

** *** ***** (~*.* lux), ** ***** **** letters ** *** **** chart, ** **** ** the ***** ** *** chart ****** **** **** defined **** ***** *** Hikvision ****. ***** *** Axis **.* **** ** the ********* ******, *** scene *******, *** ***** are *******. **** ******* in ** ******* ******** image ***** *** ********* ******/****, but ****** ** ** increase ** ****** *******, as **** *** ****** are ** **** ***********.

Open ***** *******

** ****** *** ******* in *** **** ***** outdoor ******, ***** ******* across *** **** ** a ********/**** *****, ***** ~0.01 **. ** ******* lights **** *** ******** were ****** **.

** **** *****, *** Q1635 *** ****** **** ***** 1/2" ****** ****** ******, though ******* ********* *** roughly ******* *** ** the ********* ***** ** other ******* *** ******** of *** ********* ****. It ** ******* ******** than *** */*" *****, ***** produces ** ***** ** the *******.

******** ***, *** ***** to ~**. ******* *** ******* here, **** **** ******, Hikvision *********, *** ****** details *********** *******.

***** *** *****

******* ******* **** ***** mode ** *****, *********** decreases ** *** ******* *** the ********* *********** *****. In **** ******, *** increased ***** ********** ***** our *******'* ******* ******* against *** ***** ***** background.

Parking *** *******

**** ** ****** *** cameras ** * *** parking ***, ~*.*** ** target, ********** ******, **** here:

*** ** *** ********** bright ******** ** **** scene, *********** ** *** cameras ** ******, **** the ******* *** ***** easily ************ ** *** cameras ** * ~**' HFOV.

****** *********** ****** **** pronounced **** *** *** is ******* ** ~**', though ******** *********** ** again **** ****** **** in ***** ***** ******. In **** *****, **** the **** ******* ** not **** *** *** chart, ***** *** ***** cameras ****** ***, ** varying ******, **** *** 1/2" ******* ***-**** ***** worst.

***** *** *****

******* *** ******* **** color ****, *********** ** similar ** **********. ****** in **** ***** **** not ** ********** ** in *** ***** *****, with *** ***** ***** background. **** ******* **** black, *****, *****, ***., in **** ***.

Indoor ***********

*******, ** ~*.** ***, results *** *******. ********* Darkfighter ******** *** ******** images ** **** *****, followed ** *** ***** starlight *****. **** ** the ***** ******* ******** usable ******* ** *** test *******.

** **** *** *****, below ~*.** ***, **** the ********* *** ***** cameras ******* ****** ** all.

 

Bandwidth ***********

*** ***** *** *** lowest *** ***** ******** of *** ****** ** this ****, *** ** Zipstream, ********* **** **** 1 **/* ** **** ~0.3 *** *** ~*.** lux. *** **** ****** 1/2" ***** *** ***** model *** *** ********* DarkFighter ****, ***** ******** their *.***+ ***** ***** (similar ** *********). **** that **** ******* ****** spiked ************ ** **** low *****, ****** *** bitrate ** ***** **** (~32 **/* *** ** Mb/s ** *** **** and ****, ************). 

Test **********

******* **** ****** ***** default ********, **** *** following ******** ******* ** all *******:

  • ******* ***** *** *** to * ******* ** 1/30s
  • *.*** **** *******, ** FPS, ~** ************ *** used

*** ********* ********/******** ******** were ****:

  • **** *****: *.**.*.*
  • **** *****: *.**.*.*
  • ***** ***-*******: *.***.******* **.*.*, build: ****-**-**
  • ********* **-***********-*: **.*.*
  • ******* ***-****: *.*********
  • ******* ***-****: *.*********
  • ***********: *.*.*.***** 

Comments (19)

[Note: Poster is from Raytec Canada, an external illuminator manufacturer]

all the images suck, this clearly defines the fact that outdoors in a very low light scene every camera requires light, be it IR or White light! no one should accept these images, even in the context of price as a driver.

Agree: 3
Disagree: 3
Informative
Unhelpful: 3
Funny

"all the images suck"

Beauty is in the eye of the beholder...

That said, a few years ago, even in moderately dark conditions, most surveillance cameras where pitch black. It's gone from that to an image showing quite a bit of details.

Agree: 3
Disagree
Informative
Unhelpful
Funny

Paolo, you wouldn't happen to be the same Paolo Polano from RAYTEC Canada by any chance?

If so, now I can see where you're coming from when you say

this clearly defines the fact that outdoors in a very low light scene every camera requires light, be it IR or White light!

and

no one should accept these images

and of course

all the images suck

Next time, maybe mention it up front or try a little subtlety at least?

Agree: 2
Disagree
Informative
Unhelpful
Funny

I did not hide my name like you. at the end of the day I stand by what I said. quality images need light, plain and simple.

if you do not agree that is fine also.

Agree
Disagree
Informative
Unhelpful
Funny

Paolo, IPVM requires disclosing relevant affiliations when criticizing or advocating products. I've added that to your original comment.

I also think the tone of your comment and failure to disclose up front is a blemish on Raytec and I'd advise you to review this tactic with David Lambert.

Agree
Disagree
Informative
Unhelpful
Funny

I'd advise you to review this tactic with David Lambert.

Good call. He's got quite lot of experience with it himself. ;)

Agree
Disagree
Informative
Unhelpful: 1
Funny

Let's let bygones be bygones, ok? :)

Anyway, there's a new discussion needing your input:

Does Your IR Illumination Advertise Your Blind Spots To Criminals?

Agree
Disagree
Informative
Unhelpful
Funny

IR is OK but lack of color information, where the low light cameras could help a lot expecially in moderately dark conditions with low speed objects. For most CCTV Customers, color information is always valuable.

Agree
Disagree
Informative
Unhelpful
Funny

This image, taken at 0.05 lux does not suck. Nor is it washed out like IR can do.

BTW, No one is accepting these cameras because of price, they are the more expensive models from every manufacturer.

Its odd that you would be so disappointed in these images anyway since, as John pointed out they were far worse just recently.

Maybe another f-stop or two will help you see the light? ;)

Agree
Disagree: 1
Informative
Unhelpful: 1
Funny

On second look, I have to agree; these images wouldn't be hurt by a little external illumination.

And, come to think of it my Axton Smart-AT has never let me down, illuminating my courtyard in a way an integrated IR camera would be jealous of.

Just a tip also, you might automatically think that IR illumination = RAYTEC, but there are other choices and possibly better values out there, especially when your customer can't stomach spending more on LEDS than the camera!

Agree
Disagree
Informative
Unhelpful: 1
Funny

I have had terrific success with Sony cameras in low light scenarios, and given that the SNC-VB635 compares identically to Axis' 1635 MSRP-wise, I'd love to see how it compares.

Agree
Disagree: 1
Informative
Unhelpful
Funny: 1

Thank you, this was very interesting!

Agree
Disagree
Informative
Unhelpful
Funny

In very low light, below ~0.01 lux, only the Hikvision and Dahua cameras produce images at all.

Well, there you have it!

With the latest and greatest low-light technology available on the market, ~0.01 lux images are at the very edge of usability.

Shame on all the manufacturers touting .001 and .0001 lux, what are they smoking?

Agree: 2
Disagree
Informative
Unhelpful: 1
Funny

It would be interesting to see a test of low light cameras on moving vehicles to see if license plates can be read both approaching a camera and leaving a camera. We have 4 Axis 1765LE cameras and we had to narrow down the FoV quite a bit and adjust other settings to be able to read a license plate at night. We had issues with washout and blurring but after some adjustment we can read probably about 70% of plates although the image is very dark. It would be good so see what other cameras can do. Once the light goes below a certain level night mode kicks in and it zooms into a predefined area. The image gets pretty dark, but the plates can be read. If we did not do that we had a very good color image of the overall scene, but had zero chance of reading plates.

Agree: 1
Disagree
Informative
Unhelpful
Funny

Larry,

For sure, the shutter speed would need to be shorter / faster in that scenario. We test with 1/30s shutter but with a moving vehicle, getting a clear plate would require 1/100s or 1/200s, etc. (depending on just how fast it is). And with the faster shutter speed, that means less light and a darker image.

To that end, at night, you really need IR for getting license plates on moving vehicles. Related: License Plate Shootout

Agree
Disagree
Informative
Unhelpful
Funny

John,

Yeah, we are using IR and we adjusted the shutter speed faster which would cause less light. I will look at the shoot out.

Agree
Disagree
Informative
Unhelpful
Funny

Paulo, not every scene/situation permits active illumination. In some real-world situations, active illumination is tactically unsound and may be lethal. Please do not assume that your perspective, experience, or professional self-interest encompasses everyone else's needs.

IPVM is one of the few groups performing objective camera comparison tests, especially those at the fringe of usability. Their work has been tremendously helpful in characterizing total system performance, insuring graceful video system degradation, establishing worst-case performance, and documenting relative performance differences between camera generations/performance evolution.

Agree: 1
Disagree: 1
Informative
Unhelpful
Funny

How does the Axis Q-Series':

-Lightfinder

-WDR

-bandwidth/Zipstream findings

Match up against Sony's EM/VM series?

Agree
Disagree
Informative
Unhelpful
Funny

We haven't tested the 1/2" Sony VB model, so I'm not sure how it compares to the Axis Q1635, but:

The Sony VB630 was moderately better in low light than the Axis Q1615, both 1/3" 1080p models.

Sony's WDR was similar to Axis' Forensic Capture (their latest WDR variant).

Bandwidth isn't a competition anymore. Axis wins with Zipstream. As would a lot of smart CODEC cameras vs. cameras without. Sony introduced what they call "Intelligent Coding" in their 4K model but I don't think that's filtered to the rest of the line yet.

Agree
Disagree
Informative: 1
Unhelpful
Funny
Read this IPVM report for free.

This article is part of IPVM's 6,887 reports, 921 tests and is only available to members. To get a one-time preview of our work, enter your work email to access the full article.

Already a member? Login here | Join now
Loading Related Reports