CSD’s New exacqVision Pro v5
Posted by Security Electronics and Networks | @Articles Product Reviews | December 2, 2012, 8:00am AEDT
I CHECKED out the Exacq Technologies exacqVision Pro v5 with Ilya Malkin down at CSD’s Alexandria office. Driving through an exacqVision client installed on Malkin’s laptop over a mobile network, we viewed a local system comprising a megapixel camera installed downstairs on a screen in the boardroom, which gave us a pretty good feel for the operational nature of this VMS.
Before we launch into the product, it’s worth bearing in mind that Exacq Technologies is an American company that was founded back in 2002 that specializes in open architecture NVR solutions allowing both remote and centralized monitoring. CSD has been distributing exacqVision for about 12 months and having spent some time getting to know the system, it looks to me like they’re onto a winner.
As a network video pioneer, Exacq Technologies has a good handle on the demands of high resolution video solutions. The company’s systems are supplied either preconfigured on NVRs or as NVR software. Operationally, each server is expandable up to 64 IP cameras (its 24 IP cameras on EL-Series appliances) and up to 128 IP cameras on Z-Series servers.
These servers are spread out across an enterprise and remote clients with a titchy 13MB footprint pull them together onto single screens on authorised remote workstations as if they were in the next room. Integration is possible too, with leading access control systems, retail analytics, iSCSI storage, wireless networking and video analytics solutions supported.
As well as this ability to germinate a control room on any authorised workstation, to my mind the neatest aspect of Exacq’s technology is that it doesn’t re-encode H.264 video streams as they arrive – it stores them just as they are. This pushes the work load to the edge and speeds up processing, allowing higher performance and less latency across the myriad connections of an enterprise network.
Something else I like is the way the system can display a single megapixel image as multiple region-of-interest scenes in tiles on one screen, recalling each group of scenes on playback as if they were live and all the while recording just a single megapixel stream. It sounds a bit complicated but it’s as simple as it is powerful.
According to Malkin, exacqVision has some other major advantages over the competition. For a start, it’s multiplatform and its server can run on Windows or Linux and clients can run on Windows, or Linux or Mac.
“There’s also a web browser client that supports any browser including Java screen support so users can use mobile devices to go into the system and view cameras live, play them back and zoom in, do everything you want to do with megapixel cameras,” Malkin explains.
“exacqVision also supports camera-based video analytics from a number of different vendors – and every analytic from the supported cameras can be utilised as a feature.
“It’s an open architecture VMS that supports nearly 1000 IP camera types as well as all ONVIF cameras. exacqVision also has high level integration into access control systems – including our new Integriti platform and Insight.
“It’s an open architecture VMS that supports nearly 1000 IP camera types as well as all ONVIF cameras. exacqVision also has high level integration into access control systems – including our new Integriti platform and Insight”
“Along with a list of cameras supported, there’s a list on the website of all the things exacqVision combined with access control can do, including stream video and create events. It’s a full high level interface that’s very nicely executed and developed by engineers from Inner Range using the exacqVision SDK.”
As Malkin sets up the system, he explains the nature of the exacqVision, including that intriguing characteristic of streamlining its operations by minimising the effort designated to decoding and then re-encoding video signals before storing them.
“exacqVision is not resource hungry,” Malkin says. “Instead the whole idea of exacqVision is to utilise camera abilities and functionality as much as possible. Our competitors get H.264 video streams and then they re-encode them in their own version of H.264 and that requires a huge amount of processing power. In comparison, exacqvision gets a video stream from a camera and stores it just as it is. If it’s H.264 and ONVIF-compliant it’s perfectly fine to store.
“Re-encoding certainly has its advantages with post-recording processing but if you are going to record 32 channels in real time on systems that utilise re-encoding you have to have an expensive server with dual-core Xeon processors and a motherboard with a huge amount of RAM. In an enterprise system there will need to be multiple high end servers.
“But with exacqVision we can run up to 64 IP cameras on one server with a single i7 processor running Win7,” Malkin explains. “The system also supports multi-monitor clients so more than one client can be running at a time on the same system. And thanks to the efficiency of the system, the number of clients that can be connected is limited only by bandwidth.
“The system’s powerful capability in this area relates directly to its ability to manage video streams without decoding and re-encoding. exacqVision just grabs and stores them based on camera events and out across the enterprise network clients do the hard work of decompressing the H.264 and putting it onto monitors.”
Taking this edge capability to another level, the system supports edge recording for compatible cameras from manufacturers like Axis and ISD, allowing decentralising of system storage. With edge recording, exacqVision runs a mini server on the SD card of every camera, recording and analysing everything off that camera.
“Edge recording has its advantages,” Malkin says. “Every camera is a mini system and does its own thing. And if one camera loses storage you only lose one camera’s storage.”
The powerful, easy-to-operate client user interface which we use to steer the system installs on any standard computer running Windows, Linux or Mac OS X. It connects to multiple exacqVision Hybrid (analogue and IP cameras) and IP Camera Servers creating a virtually unlimited amount of cameras. Malkin tells me he’s not fond of this ‘unlimited’ label and very correctly points out that the governor of camera numbers is bandwidth.
It’s a very familiar management screen we are looking at, I say, as we settle down behind the client to drive the system. There’s a camera tree with cameras labelled by type or location on the left, a tool bar of screen and zoom options at the top and the rest of the space is taken up by a split screen.
When you get into search, searchable cameras are called up in the left hand directory and tabs underneath allow you to search by maps, groups, views, events or cameras. A comprehensive timeline appears underneath the viewing tiles. With a single click you can pull a scene to full screen.
In enterprise applications –there’s single-screen, real-time view of all cameras in the enterprise system. These can be sorted by connection, recording status, IP address, or firmware revision. To manage a camera you just double-click for that camera’s setup screen or right click any camera for live or instant playback via exacqReplay.
“Yes – it’s very easy to operate – not complicated at all and it’s not a busy GUI. It’s user friendly and the user interface is intuitive,” says Malkin. “Something worth point out is that the system has a very special digital PTZ functionality including regions of interest.
“It actually creates views of the same megapixel camera and stores them so they can be recalled and played back in the same configuration they are displayed live. What this means is that in one single megapixel stream I can get multiple different views from one camera but view them as if they are separate images – that means no zooming or panning into regions of interest during investigations.”
Malkin shows me how setup of these region-of-interest groups is achieved from a single megapixel camera.
“It’s very simple stuff, really. I go to the camera with the cursor, grab a megapixel stream, 1, 2, 3, 4 times. Say I want to zoom to this particular area here, and this particular area here, and this area here. I zoom in on an area of the scene in the relevant scene window and I just click ‘apply’ to save the views. Later on I can change the separate ROI views in exactly the same way.”
To illustrate the point, Malkin goes through the process of setup using the camera downstairs in the showroom. Bear in mind we’re running over a mobile network for this test and performance is typically 3G. I’m actually pleased we’re using a wireless hotspot for the demo because it gives a sense of an enterprise solution.
“Ok – that’s the main GUI and I’ve created a particular group of views from a single camera downstairs in the showroom,” Malkin tells me. “That one megapixel camera shows us inside the front door, outside in the carpark and all these other areas in the showroom. From the point of view of a user or integrator this means one 5MP camera facing a building can cover a main entrance, a dock entrance, a car park, and multiple other areas for viewing and recording.
“As a manager or operator this means if I want to look at the activity that’s taken place over a 2-hour period in the car park, I don’t have to zoom in on the carpark – the camera view is recorded complete but is split into the area of interest views so I can view the areas of interest on playback as I view them live.”
The way the client works is that it links to multiple exacqVision servers to create scalable networks that expand pretty much endlessly. At the heart of the operation is multi-monitor client display & video wall with the ability to drag and drop cameras and maps between monitors, which is awfully cool. There’s easy to use multi-camera search, playback and export and there are also video, audio and POS export features.
There’s a graphical search display for cameras and recorded audio and a thumbnail search for instant visual assessment and video retrieval. The strength of this capability it that it’s perfect for bandwidth-poor remote connections – and for mobile applications too, given the annoyance of modern contention rates in Australian networks, particularly at peak times.
exacqVision mapping integrates seamlessly with video streams. To make this work you just add any number of maps from any image source, place cameras and alarms on the maps. Neat too, is the ability to built user-configurable triggers on maps for controlling access control systems. Once the map is setup you can then view interactive activity on map.
According to Malkin, exacqVision is an ideal system for one or 2 cameras or for huge enterprise systems. This is a big call but the nature of Exacq’s edge-leaning technology is so sleek that once you get your head around it you can see just how capable exacqVision can be.
“Enterprise features include managing all users from a single system location – absolutely everything can be managed from the one point,” Malkin says. “If there are 20 systems with 50 users, some users are allowed to see some camera views from one part of the system, while other users can only see other cameras.
“There’s also full integration into active directory which is required by all high end projects so as to allow system administrators to manage all linked systems. exacqVision also has multi-level mapping so if there is building of 8 storeys, every storey can be integrated into the main map and all the floors and wings of the building can be displayed.”
Another nice feature is an optional enterprise health manager server. The server runs a separate SQL database so it keeps monitoring the whole enterprise and running diagnostics about video loss of any camera, loss of server connection, HDD failure, across the whole enterprise, reporting system problems live.
exacqVision is right on up the functionality du jour and supports iPhone, iPad, Android, Kindle Fire and Blackberry Playbook browsers. Recently, multi-streaming was added and these streams can be configured in different ways depending on the importance of the events. For instance, alarm events will be recorded at a higher frame rate, at higher resolution with less compression. That’s cool when you’re viewing scenes on a mobile device.
So – can the system be installed on off-the-shelf servers? I ask. What’s the process of building exacqVision solutions from an integrator’s point of view? And what about licensing – how does that work?
“We generally provide end-to-end solutions and we prefer to sell our server with our warranty that’s specified and dedicated to handling this sort of a role,” Malkin tells me.
“We have a number of different types of servers with licenses pre-installed. These include the exacqVision A-Series Hybrid and IP servers, enterprise-level Z-Series Hybrid and IP servers and the EL-S/EL-SR hybrid and the EL-Series IP embedded appliance servers.
“Our licensing is very simple you pay licenses only for the number of cameras you want to connect and it’s not connected to a camera MAC address. If you want to upgrade a camera, you just replace the camera – you don’t have to pay another license.”
According to Malkin, from an integrator’s point of view, setup is made easy because the system searches and enrols cameras by itself.
“I can add a camera simply by choosing a manufacturer and selecting the relevant camera on the list. The system will then find the camera on the network and enrol it.”
For cameras not registered, there’s RTSP (real time streaming protocol) if you need it. And ultimate megapixel performance is offered thanks to support for Arecont’s monster 20MP cameras.
“Something else that strong about the system also has full audit trail so users can check who has done what on the system, see who has logged in and logged out of cameras, who has configured camera views, every single thing that has been done on the system can be monitored and checked,” Malkin says.
“And importantly, Exacq Technologies provides quarterly updates for exacqVision that include new features and the addition of newly compatible cameras.”