Testing Avigilon's H3 WDR Camera

Published Feb 14, 2013 05:00 AM

Avigilon has long marketed itself as delivering 'the best evidence' but until recently they did not have any WDR cameras. However, their H3 release introduced WDR up to 3MP.

Here are our key findings from our tests:

  • Solid WDR performance compared to top WDR cameras.
  • Average low light performance compared to other HD WDR cameras.
  • Daytime bandwidth consumption at or below competitive cameras.
  • Nighttime bandwidth consumption spiked, among the highest in our test.
  • Default compression set well below other manufacturers. Consider increasing to lower bandwidth while maintaining competitive image quality.
  • New 2.42 firmware fixed multiple issues with previous version, including autofocus malfunctions, as well as color and saturation issues with WDR turned on.
  • Price: The Avigilon H3 3MP has an MSRP of $835. This is at or below street price of many competitors, such as Axis, Bosch, Panasonic, and Sony, which range from ~$800 to $1,100 online. This makes it one of the lowest cost 3MP WDR cameras available from a major manufacturer.

Recommendations

Avigilon's H3 series 3MP camera is a competitively priced for its featureset, including auto focus and solid WDR performance. However, those looking for superior low light performance, as well as more advanced features such as on-board storage or analytics should consider other offerings.

For those currently predominantly selling Avigilon, the 3MP WDR model represents a great advance in WDR performance, and a modest increase in low light imaging compared to older models.

Configuration Overview

This screencast overviews key camera configuration and settings.

WDR Performance

We tested the H3's WDR performance against five competitors in our 2013 WDR Shootout. We found that performance was solid in all of our test scenes.

These example images show performance in a garage scene, with an overhead door open with strong sunlight behind the subject (click for larger versions). In this comparison, we can see that the subject's face and the chart are well-lit, and details outside are visible, as well. The H3 is only among the top performers in this scene.

With the subject standing in the darker area next to the door, the chart and subject remain quite visible, more so than in the Bosch, Panasonic, and Sony models, and only moderately darker than the only 3MP competitor, the Arecont AV3116DN.

 

Low Light Performance

In this indoor nighttime scene, below 1 lux, Avigilon produces one of the dimmest images among the cameras tested. Note that this comparison shows the Arecont AV3116 with binning turned on. Avigilon outperforms only the Sony CH240 and Panasonic WV-SP509.

Indoor Performance

In our indoor, well-lit test scene (~300 lx), the H3 performs well. Colors are fairly true, but slightly oversaturated, as can be seen in the subject's face. Whites also display a slight blue tint. Neither of these issues had a major impact on image quality, however, and could potentially be remedied by adjusting camera settings if need be.

Image Quality in Avigilon vs. 3rd Party VMS

For those using an end-to-end Avigilon system, the H3's image quality will differ, due to Avigilon Control Center's automatic image adjustment features. This adjusts gamma, white, and black levels automatically with the intent of optimizing in the image. In our tests, we found modest improvement to the H3's image during daytime, essentially amounting to a slight white balance adjustment:

However, ACC provided a marked improvement in our low light scene, brightening the image enough that our test chart may be seen, not possible in Exacq or other clients.

Bandwidth Consumption

In order to test bandwidth, we set all VBR cameras to their default settings, but increased the cap to 12 Mb/s, in order to provide substantial headroom for the camera to adjust bandwidth accordingly. Note that only Sony's CH240 did not include VBR streaming, so we set its CBR bitrate to 8 Mb/s. In our daytime scene, the Avigilon H3 consumed less bandwidth than most other 1080p/3MP cameras in our test, aside from the Bosch NBN-932V.

  • Avigilon H3 3MP: ~3 Mb/s
  • Arecont AV3116DNv1: 8-10.5 Mb/s
  • Axis Q1604: 1.4-1.6 Mb/s
  • Bosch NBN-932V: 700-1,000 Kb/s
  • Panasonic WV-SP509: ~8 Mb/s
  • Sony SNC-CH240: 8 Mb/s CBR

However, at night, bandwidth spiked, higher than all cameras aside from the Arecont AV3116DN. However, note that the Axis Q1604 consumed around 9 Mb/s at 1.3MP resolution:

  • Avigilon H3 3MP: 10-12 Mb/s
  • Arecont AV3116DNv1: 9-12 Mb/s
  • Axis Q1604: ~9 Mb/s at 1.3MP
  • Bosch NBN-932V: 4.5-5 Mb/s
  • Panasonic WV-SP509: 2.5-3.5 Mb/s
  • Sony SNC-CH240: 8 Mb/s CBR

Very Low Default Compression

One factor contributing to Avigilon's higher bandwidth consumption is their default compression settings, low relative to many competitive cameras. Avigilon's H3 series defaults to a quality level of "6" in their web interface, which tests determine equates to a quantization level of 24, while most other manufacturers default to somewhere around 30. Left defaulted, the H3's stream jumped to 7-8 Mb/s in our daytime indoor test scene, compared to 2.5-3 Mb/s when adjusted to "15", roughly 30 on the quantization scale, with minimal impact to visual clarity. Nightime bandwidth was impacted much less, with consumption dropping from 10-12 to about 9.5-10.5, meaning aggressive gain control is likely a bigger contributing factor than compression in low light scenes. Users should be aware of this default setting and adjust compression accordingly for their scene.

Comments are shown for subscribers only. Login or Join