The right equipment for reliable AI video analytics
Cameras, recorders, networking and Edge AI computers: VIDEO.MA reuses your existing infrastructure when it is compatible and proposes new equipment when the analysis requires it. Every choice follows a technical audit of your site.
Every project starts with a technical audit. Based on its findings, VIDEO.MA builds on your existing infrastructure or proposes new Smart Vision equipment — and often a combination of both.
01
Reuse your compatible infrastructure
When your cameras and network allow it, adding an Edge AI computer is enough to turn your CCTV system into an analytics tool.
Check of camera resolution and positioning against the target analytics
Assessment of network capacity and recorder performance
Only an Edge AI box is added, with no blanket replacement
Targeted recommendations if a camera needs repositioning or replacing
02
New VIDEO.MA Smart Vision equipment
When existing infrastructure is missing or unsuitable, VIDEO.MA designs a complete package dedicated to video analytics.
Cameras selected and positioned specifically for analytics
Edge AI computer sized to the number of streams
Networking, PoE switches and storage suited to the site
On-site installation, configuration and calibration
Indicator validation during a pilot phase
IP cameras
Network cameras delivering a digital video stream that analytics models can process. They are the data source for any Smart Vision solution.
Typical use cases
— People counting at entrances
— Occupancy and zone analysis
— Replacing analogue cameras that are hard to analyse
What we assess
— Resolution matched to subject distance and the area to cover
— ONVIF compatibility and RTSP stream access
— H.264 / H.265 compression and bitrate the network can handle
— Low-light performance and infrared range
Dome cameras
Compact, discreet ceiling-mounted cameras suited to indoor spaces. An overhead view reduces occlusion and makes counting easier.
Typical use cases
— Top-down counting above an entrance
— Covering sales floors, rooms and lobbies
— Monitoring queues at checkouts or service desks
What we assess
— Available ceiling height and viewing angle
— Field of view (fixed or varifocal lens)
— Vandal resistance (IK rating) where needed
— PoE power and visual fit with the premises
Bullet cameras
Directional cylindrical cameras, usually mounted outdoors or high up, suited to long distances and wide areas.
Typical use cases
— Counting vehicles at access points and in lanes
— Monitoring docks, yards and car parks
— Covering outdoor perimeters
What we assess
— IP rating suited to dust, rain and heat
— Lens focal length matched to subject distance
— Infrared range and backlight handling (WDR)
— Mounting conditions and orientation relative to the sun
PTZ cameras
Motorised cameras that can pan, tilt and zoom. They are useful for monitoring large areas but less suited to analytics that require fixed framing.
Typical use cases
— Visual supervision of large sites
— Verifying a situation after an anomaly signal
— Covering wide areas with preset positions
What we assess
— Compatibility with analytics (fixed presets during analysis)
— Optical zoom range relative to distances
— ONVIF control and integration with the existing system
— Required PoE power budget
NVR / DVR recorders
Network video recorders (NVR) for IP cameras or digital video recorders (DVR) for analogue cameras. They handle recording and, depending on the model, can relay streams to the analytics engine.
Typical use cases
— Recording and archiving surveillance footage
— Accessing existing streams via RTSP for AI analysis
— Centralising multiple cameras on a site
What we assess
— Number of channels and supported incoming bitrate
— Ability to relay streams over RTSP / ONVIF
— Storage capacity relative to the retention period
— Compatibility with installed cameras
PoE switches
Network switches that power cameras and devices over the Ethernet cable (Power over Ethernet), simplifying installation and reducing cabling.
Typical use cases
— Powering and connecting IP cameras
— Segmenting the video network
— Extending an existing installation
What we assess
— Total PoE budget relative to device consumption
— Port count and uplink bandwidth
— Managed features (VLAN, QoS) to isolate video traffic
— Cable runs between the switch and the cameras
Edge AI computers
Compact on-site computers that run video analytics models close to the cameras. Only aggregated data leaves the site, which reduces bandwidth needs and privacy risks.
Typical use cases
— Real-time analysis of streams from several cameras
— Adding intelligence to an existing CCTV system
— Operating with limited Internet connectivity
What we assess
— Compute capacity relative to the number of streams and analysis frame rate
— Compatibility with the selected models and inference tools
— Installation conditions: temperature, ventilation, power
— Remote monitoring and update capabilities
AI accelerators
Specialised compute modules or cards (GPU, NPU) that speed up neural network inference, to process more streams or run more demanding models.
Typical use cases
— Scaling up to a larger number of cameras
— Running more accurate detection models
— Consolidating analysis on a local server
What we assess
— Compatibility with the host platform and inference software
— Performance gain measured on the models actually used
— Power consumption and heat dissipation
Network equipment
Routers, firewalls, point-to-point wireless links and Internet access needed to connect cameras, edge computers and servers reliably and securely.
— Separating the video network from the office network
What we assess
— Available bandwidth relative to the number of streams
— Access security (VPN, firewall, segmentation)
— Link reliability and failover options
— Distances and building constraints
Storage equipment
Surveillance-grade drives, NAS units and storage servers sized for recording video streams and retaining analytics data.
Typical use cases
— Retaining video recordings for the permitted period
— Storing aggregated analytics data
— Building sample sets to validate a pilot
What we assess
— Number of retention days and camera bitrates
— Drives designed for continuous recording
— Redundancy (RAID) and backup
— Retention periods compliant with Law 09-08 and CNDP authorisations
Complete Smart Vision kits
Coherent packages combining cameras positioned for analytics, an Edge AI computer and networking equipment, configured for a specific use case. Their composition is defined project by project.
Typical use cases
— Launching a pilot on one site
— Rolling out the same use case across several sites
— Sites with no usable video installation
What we assess
— Target use case and indicators
— Number of measurement points and site layout
— Installation and network constraints
— Plans to extend to other areas or sites
Equipment is selected project by project, following a technical audit of the site. Brands and models are proposed in the quotation based on requirements, compatibility and availability. VIDEO.MA does not publish a catalogue or public prices.
Frequently asked questions
Can I use my current CCTV cameras?
Often, yes. If your cameras are IP cameras accessible over RTSP or ONVIF, with suitable resolution and viewing angle, adding an Edge AI computer is usually enough. The compatibility audit checks this camera by camera.
Are my analogue cameras compatible?
Streams from a DVR can sometimes be used, but image quality is often too low for reliable analysis. In that case, VIDEO.MA recommends replacing only the cameras located at measurement points.
Why process video on site rather than in the cloud?
Edge AI processing reduces bandwidth requirements, works even with limited Internet connectivity and avoids sending images off site: only aggregated, anonymous indicators are transmitted.
Which equipment brands do you offer?
VIDEO.MA is not tied to a single brand. Equipment is chosen after the audit according to your technical constraints and budget, and brands and models are specified in the quotation.
Do you handle installation and maintenance?
Yes. VIDEO.MA takes care of installation, configuration, analytics calibration and equipment maintenance, under the terms defined for each project.
Ready to turn your cameras into business intelligence?
Tell us about your business: we’ll show you what your cameras can measure, backed by an honest feasibility study.