Worlds First Review - Dahua DH-IPC-HFW5449T1-ASE+D2 - Dual Lens Full Color 2.0

Wildcat_1

Known around here
Dec 9, 2018
2,129
6,092
US
Intro

I’m pleased to announce we have another worlds first review to kick off, this time on the new HFW5449T1. So whats better than 1 camera (lens), well 2 of course. This camera makes some bold claims of Full Color regardless of low light environments utilizing 2 lenses. Lets jump into it but first a big THANK YOU as always to @EMPIRETECANDY for providing this unit in exchange for a fair, honest, unbiased and no holds barred review.​
I’ll attempt to answer any and all comments as always BUT please keep those to the main topic of this thread. Post 2 - Pics & Videos, Post 3 - Special Features / Functions / Tips & Tricks, Post 4 - Bug List (new), Post 5 - Wrap Up​
Lets go….​


The Camera

Well for starters the camera model numbers just seem to keep getting longer but luckily in this case the body/footprint of the camera is not. In fact this camera adopts the same new style of bullet cam build as the recent IPC-Color4K-X. Remember this is a dual sensor / dual lens unit implemented by a 1/1.8” for the main and 1/2.8” for the secondary lens. The primary lens (1/1.8” sensor) features the f1.0 aperture and the secondary (1/2,8”) features an f1.6 aperture. Let me also be clear in pointing out that you configure the main cam (primary lens) and the cam combines the input from both lenses / sensors as required. i.e. NOT dual channels as in other cams I’ve reviewed. This also means that each lens/sensor is NOT directly addressable for you as an end user. While you can adjust settings and configuration that ultimately impact 1 or both lenses on the backend, you address the config as a camera and not separate lenses/sensors. Therefore the real power of a unit like this relies heavily on its AI algorithms and how well it utilizes both sensors to deliver the end result.​
The camera comes in standard fixed length options, 2.8, 3.6, 6mm and I will be testing the 3.6mm version in this review.​
This camera has some clear targets for Dahua, specifically in reducing light pollution by utilizing 2 sensor/lens combo to ensure that any and all color information is captured from either lens then spliced together to allow for less use of light on scene. More on the ‘splicing’ as I go through the review. Therefore a primary use case here are for those that either don’t have, can’t have, don’t want too much light in their FOC BUT still need decent captures. Talking about light, you do still have options with this camera, quite a few. It features:​
  • IR Mode - Auto / Manual (1 slider, no control over near and far)
  • White Light - Auto / Manual
  • Soft Light - Auto / Manual. This one is a combination of IR + white light combined
I do want to preface / caveat up front that this camera is definitely ‘unique’ and as you will see basically creates a fused image between the lenses. As I will show in my night testing, this produces a very interesting end result which is unlike most video people here will be used to. Bear with me though, I show how this works in some creative ways ;)


Specifications


Shipping FW Version - Thought I would start adding this for reference


2021-08-05

Mounting Options

The mount options for the HFW5449T1 are:​

  • PFA122 - 3 hole junction box
  • PFA130E - Circular junction box
  • PFA152E - Pole Mount Adapter
Remember my tip about using PFA121 & PFA122 junction boxes and quick switching the plates (4 hole vs 3 hole), very useful here too. Just remember this one needs the PFA122 (3 hole plate)

Wizmind Series 5 Features

This camera is part of the Dahua Wizmind range so not only gets IVS based object filtering (Human, Vehicle, Non-Vehicle) as well as Face Detection, Heat Mapping, People Counting but also SMD 3.0. As this camera features the newer FW you will see support for Smart Plan schedules + the new Sensitivity and ‘inside’ options for IVS triggers introduced with the 4K full color cams.​


Camera Details

Image Sensor - 1/1.8” + 1/2.8”​
Effective Pixels - 2688(H) x 1520(V), 4 Megapixels​
RAM: 1GB​
ROM: 128MB​
Electronic Shutter Speed - 1/3s~1/100,000s​
Minimum Illumination
Color: 0.0002 Lux @F1.0​
Illumination Distance (utilizing the additional LED options) - Up to 50m (164.04 ft)​
IR On/Off Control - Auto/Manual​
Illuminator LEDs - 4 IR, 2 Warm Light​


Lens Information

Focal Length (Fixed)
2.8, 3.6, 6mm​
Max. Aperture
Primary Lens 1 - F1.0 (constant)​
Secondary Lens 2 - F1.6 (constant)​
Field of View
2.8mm
H: 97°​
V: 48°​
D: 115°​
3.6mm
H: 75°​
V: 39°​
D: 87°​
6mm
H: 47°​
V: 25°​
D: 56°​
Close Focus Distance
2.8mm - 2m (6.6ft)​
3.6mm - 2m (6.6ft)​
6mm - 2.5m (8.2ft)​
DORI Distance
2.8
Detect - 58.2m (190.9ft)​
Observe - 23.3m (76.4ft)​
Recognize - 11.6m (38.1ft)​
Identify - 5.8m (19ft)​
3.6
Detect - 70.3m (230.6ft)​
Observe - 28,1m (92.2ft)​
Recognize - 14.1m (46.3ft)​
Identify - 7.0m (22.3ft)​
6.0
Detect - 116.1m (380.9ft)​
Observe - 46.5m (152.6ft)​
Recognize - 23.2m (76.1ft)​
Identify - 11.6m (38.1ft)​
Audio Information
Built In Mic + Speaker​
 
Last edited:
Post 2 - Pics & Videos

Comparison Pics

Some moody comparison pics courtesy of an alarm esque light in the background :). Comparisons here to 5231-Z12 & 7442-Z4 + a close up of the lens housing

5449 Line Up 1.jpg 5449 Line Up 2.jpg 5449 Close Up.jpg


Daytime Video Test
Remember this is uploaded in native resolution (2688x1520) and edited that way too, however YouTube likes to decide how it wants to align that to standard Quad HD templates so watch on YouTube and select 1440p




Night Video Test

Ok here it is without further ado, the night test. I’ve attempted to push this cam to the limits in this test, starting with additional light then reducing down to zero light. I finish by showing creatively what this camera is capable of as it searches for light, finds whatever is available and then fuzes the image from both sensors together to produce the final image.

Bear in mind the bugs I mentioned that still need addressing around alignment but hopefully you’ll see that this camera is incredibly unique in what its aiming to do, the issues it is trying to address and how this may stack up to be an additional way to monitor those darker FOV’s or those where ideally you want color + detail + usable recognition criteria.

Take a look and prepare yourself for a different perspective on night surveillance :)

As always, make sure you choose YouTube to watch this, select 1440p. Unfortunately even though I keep this in native and edit in ProRes then upload in as high a quality as possible, YouTube will always re-process and degrade the image further :( This is a longer video so I added timestamps in the description and on the video for ease of jumping about.

Lets go…………..

 
Last edited:
Post 4 - Bug List
Below are the bugs I've found and filed to Dahua along with status. I'll also include Enhancement Requests that I file as I find areas for improvement too. Will update as I move through the review. All of these I have provided data, pointers in codebase to fix, video captures showing the issues and recommendations on implementation for next FW to Dahua.

BASED ON 2021-08-05 FW
  1. IR LED's strobe when set to Auto & target is over 6ft from camera, works fine when closer than 6ft, workaround, use manual = NEEDS TO BE FIXED
  2. Splicing calibration - Expect a FW update very shortly (next day or 2) to test a fix
    • Needs manual option as Automated doesn't allow granularity of lens alignment even when using pro calibration images = ENHANCEMENT REQUEST
    • Currently leads to phenomenon where excessive halo / fringing appears when target is close. In daylight this can soften the image due to the alignment of lenses being off = NEEDS TO BE FIXED
 
Last edited:
  • Like
Reactions: Funkenflug
You didn't think I would keep you all waiting too long for the first video did you ? Will be posting this tonight as soon as YouTube finishes its re-processing dance. I compare the 5449T1 to the 5442 in all videos within this review. Hold onto your hats guys and girls though as the night time is where this camera does things a little differently ;) Stay tuned !
 
The long and short of it is that the 121 and 122 are square boxes that the lids can be swapped between which allows easy camera changes since the lid screws are accessible when the camera is mounted to it.
 
Could you post a link to a more detailed info on this please?
The faceplates for the PFA 121 & 122 (4 & 3 hole square junction boxes respectively) are interchangeable so what I guide people to consider is using these for the mounting of your cams instead of the PFA130 (circular junction box).

The benefit is in ease of mounting additional cams or replacements. As an end user or installer, if you use the PFA 12x’s (buy a couple of each), install the junction box portion itself physically at the location (remains installed at all times) then whenever you want to switch a camera out you can mount the new cam to 1 of your spare faceplates on the ground / inside.

When it comes time to install all you do is unscrew the faceplate (with the existing/old camera still mounted to it) and replace it with the new faceplate with your new cam that you attached on the ground. Removes having to mess with taking the entire cam off the plate (often times high up or with limited accessibility) and then remounting onto the plate on location. Saves a ton of time and you can switch cams out in minutes. Great when having to do this in inclement weather as well and can make multi cam installs a breeze.

HTH
 
The faceplates for the PFA 121 & 122 (4 & 3 hole square junction boxes respectively) are interchangeable so what I guide people to consider is using these for the mounting of your cams instead of the PFA130 (circular junction box).

The benefit is in ease of mounting additional cams or replacements. As an end user or installer, if you use the PFA 12x’s (buy a couple of each), install the junction box portion itself physically at the location (remains installed at all times) then whenever you want to switch a camera out you can mount the new cam to 1 of your spare faceplates on the ground / inside.

When it comes time to install all you do is unscrew the faceplate (with the existing/old camera still mounted to it) and replace it with the new faceplate with your new cam that you attached on the ground. Removes having to mess with taking the entire cam off the plate (often times high up or with limited accessibility) and then remounting onto the plate on location. Saves a ton of time and you can switch cams out in minutes. Great when having to do this in inclement weather as well and can make multi cam installs a breeze.

HTH

I think @harleyl7 has used these to good effect in his lpr install thread. See link

 
  • Like
Reactions: harleyl7
I think @harleyl7 has used these to good effect in his lpr install thread. See link


Yes, good to see he is using these and he will benefit from easy switchout in the future :)

Like I said, I keep a bunch of these junction boxes on hand at all times to do what I’ve shared above and before. Really makes installs a lot easier :)

Might as well simplify where you can and make things easy on yourself where possible.
 
Ok everyone, here is the day time video test. I'll post the video's inline so that new viewers of the thread can keep context to any comments BUT will also add to Post 2 (videos and pics post) for easy reference at the start of the thread too. Remember this is uploaded in native resolution (2688x1520) and edited that way too, however YouTube likes to decide how it wants to align that to standard Quad HD templates so watch on YouTube and select 1440p

 
@Wildcat_1 compliments on a great video as usual. I see in the text the new cam is a 3.6mm and the other comparison 5442 cam lens is varifocal at 2.7mm. Is there a reason for the difference or is it just for illustration.
 
@Wildcat_1 compliments on a great video as usual. I see in the text the new cam is a 3.6mm and the other comparison 5442 cam lens is varifocal at 2.7mm. Is there a reason for the difference or is it just for illustration.
Just for complete transparency really @Griswalduk. The vari was set at full wide, therefore 2.7 vs the fixed on this one being the 3.6 so just making people aware of the slight focal length difference of each cam as they watch them in the comparison