SquintPRO · Cities and roads
Smart city solutions with video analytics: road and sign condition from existing cameras
Smart city solutions span networks, sensors, platforms and software. This page covers one of them: video analytics that turns the cameras a city or road operator already has into checks of road and sign condition, and other conditions a camera can see. SquintPRO, the video analytics platform Interkey builds, is deployed in a smart-city project to monitor road and sign condition. Each suspected case comes with its camera, time and a video clip, for a person to review before it is referred. Other city uses described here are product capabilities to test in a pilot on your cameras.
Published 5 min read
In short
- What it adds: checks between scheduled inspections, from fixed cameras, with video evidence of each case.
- Deployed today: road and sign condition monitoring in a smart-city project.
- Other capabilities to test: vehicles in restricted areas, crowd density at public spaces and events, tracking vehicles across cameras, and searching footage by description.
- What it does not do: decide on repairs or identify people. There is no face recognition.
- Where it runs: on your infrastructure, so footage stays on your premises, or in the cloud with 3, 6 or 12 months' retention.
- How it starts: one corridor or district, one or two conditions, and written measures.
What video analytics adds to smart city solutions
Road operators and city teams usually find damaged signs, obstructions and defects through scheduled inspections and reports from the public. Fixed cameras watch the same scene all day. Video analytics turns that footage into suspected cases between inspections:
Read more
- Capture. Cameras that overlook a corridor or area are connected to SquintPRO, usually existing ones if their angle and resolution suit the condition.
- Define. Each condition is set up from SquintPRO's 35 trained detection libraries where one fits, or described in text or with a reference image, without a model-retraining project.
- Detect and record. Each suspected case is logged with camera, time, zone and condition type, plus a video clip.
- Review. A person confirms or dismisses each case, so field teams are sent only to checked cases.
- Refer. Confirmed cases go to the maintenance team or department through the route you choose. Interkey builds that connection during implementation.
- Measure. Cases are counted by area, period and condition type, to show where the same problems recur.
Which road and sign conditions suit fixed cameras
| Condition | What a camera can show | What makes it hard |
|---|---|---|
| Damaged, leaning or fallen sign | A sign at an unusual angle, bent, or on the ground | Distance, glare, and how small the sign appears in the frame |
| Sign hidden by trees or objects | Branches or objects covering the sign face | Growth is gradual, so periodic review suits it better than instant alerts |
| Debris or an object in a lane | A stationary object on the carriageway | Short-lived objects, and traffic hiding the lane |
| Vehicle stopped in a restricted area | A stationary vehicle on a hard shoulder or at a junction | Telling a breakdown from congestion |
| Damaged barrier or road furniture | Visible bending, a gap or a displaced element | Viewing angle and small deformations |
| Potholes and surface damage | Large defects in a well-lit part of the view | The shallow angle; small cracks may not be visible at all |
A condition that stays visible for hours, such as a fallen sign, is a better candidate than a small detail that needs a close look. Cameras are weakest at small surface defects, because they see the road from a distance and at a shallow angle. The table is an assessment framework: whether a condition can be detected on your corridor is something a pilot on your own cameras has to show.
Other city uses of the same platform
The same SquintPRO deployment can run more than one use case on the same cameras. The table separates what is deployed in a city project today from product capabilities that a pilot on your cameras should test.
| Use | What SquintPRO does | Status in city projects |
|---|---|---|
| Road and sign condition | Flags suspected damage or obstruction for review | Deployed in a smart-city project |
| Vehicles in restricted areas | Alerts when a vehicle is where vehicles should not be | Product capability; test in a pilot |
| Crowds at public spaces and events | Measures density and movement, with early alerts when a set threshold is exceeded | Product capability; test in a pilot |
| Following a vehicle across cameras | Tracks people and vehicles from camera to camera on a map | Product capability; test in a pilot |
| Finding an event in recorded footage | Searches by description or with a reference image | Product capability; test in a pilot |
See also crowd management and people counting.
Visual pollution and the municipal programme
The Ministry of Municipalities and Housing runs an initiative to remove the most common visual-distortion elements in Saudi cities. The elements it lists include construction waste, the appearance of waste containers, non-compliant commercial signs, damaged vehicles, graffiti on walls, and non-compliant canopies and metal sheds. The ministry carries it out through inspection campaigns, community campaigns and urban improvement projects.
Several of these elements can stay in a fixed camera's view for days. That makes them reasonable candidates to test on cameras that face the right streets. A condition can be defined in text or with a reference photo, then tested on your footage before it raises cases.
Privacy and deployment
- No face recognition. SquintPRO detects conditions, objects, people and vehicles. It does not identify who a person is.
- Number plates are captured incidentally. Road cameras show passers-by and plates. Decide with your legal team who may view footage and cases, and how long they are kept.
- On your infrastructure or in the cloud. On-premise processing keeps footage on your premises. Cloud deployment keeps recordings for 3, 6 or 12 months.
- Into your existing systems. Cases go to the video management, control-room or incident system you already run.
How a city pilot is measured
Start with one corridor or district and one or two conditions. Agree these measures in writing before go-live, and cover weekdays, weekends and night hours:
- confirmed cases as a share of all cases raised, to see how much of the review work is useful;
- what the cameras missed: conditions found by inspection on the same corridor and period that the cameras did not raise;
- time from detection to referral, and from referral to close-out;
- daily review workload, against what your team can handle.
Then decide in writing whether to extend, adjust or stop. Read the pilot scoping guide.
RelatedRead the pilot scoping guide
Pricing and delivery
SquintPRO is priced per use case. Each quote is built from the site, the number of cameras and the use cases in scope. Interkey carries out the camera survey, integration, deployment, tuning and operation from Riyadh, in Arabic and English.
Questions city teams ask
Do we need new cameras?
Not necessarily. A camera survey shows which cameras can be used as they are, which need adjusting and which would need replacing.
Does it raise work orders automatically?
It raises suspected cases with evidence. A person reviews them, and confirmed cases are referred through the route you choose. Repair decisions stay with your team.
Has SquintPRO been used in a smart city project?
Yes. It is deployed in a smart-city project to monitor road and sign condition. Project names are not published.
Can footage stay on our infrastructure?
Yes. In an on-premise deployment, analysis runs on your infrastructure and footage does not leave your premises.
Does it identify people?
No. SquintPRO does not use face recognition.
Next
Related pages
Next step
Discuss a pilot on one corridor
Tell us the corridor or district you have in mind, roughly how many cameras overlook it, and the conditions that matter most to your team. We will come back with a first view of camera suitability and a pilot scope with written measures.
Or directly
+966-11-2180999 [email protected]Tawuniya Towers, North Tower, 7th Floor, King Fahad Highway, Olaya, P.O. Box 56835, Riyadh 11564, Saudi Arabia