Mining and quarries
Slopes, benches and haul roads: ground movement that advances for weeks before it becomes a rockfall.
Rust Neural
The crack was already there before anyone saw it.
Corrosion, wear and loose anchors: things that change slowly and that nobody sees for that very reason, until it is late.
What changes slowly the eye does not see.
Somebody who walks past the same structure every day does not see it rusting. The person who comes back a year later does.
That is why condition inspections are done with photos and written records: so you can compare. And that is why they happen every six months or every year, because setting up the comparison costs money.
Between one inspection and the next there is a camera looking at that same structure twenty-four hours a day and keeping all of it.
The same photo, every day.
Because the camera does not move, today's image and the one from three months ago are identical except in what has changed. That is all it takes.
You mark which element has to be watched and the system compares. When the difference crosses a threshold, it alerts you with the two images side by side.
And for the formal inspection, instead of starting from scratch there is a series: how it looked in January, in March and in June.
What it watches.
- Corrosion, rust and wear, and how they advance over time.
- Cracks and fissures that lengthen between inspections.
- Broken components: handrails, mesh, grilles, supports.
- Anchors and bolts that shift.
- Movement and rockfalls on slopes and wall faces.
- Blockages and build-up in drains, screens and channels.
- Vegetation growing where it puts the installation at risk.
- Levels, pressure and readings from analogue instruments.
- The temperature of machinery and components, with a thermal camera.
- Leaks, drainage and flooding.
- Fire and smoke.
Lo que Rust Neural entiende
21 detecciones grabadas.
Estado de máquinas e instalaciones.
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Rust
Cada una es una grabación del sistema funcionando. Pasa el ratón por encima —o tabula hasta ella— para verla en movimiento.
A real case.
On a mountain lift, an anchor at the mid station started to show play. A normal visual inspection does not pick that up: it is millimetres.
The monthly comparison flagged it on the second reading. It was tightened during the May maintenance shutdown. The alternative was finding it at the annual inspection in October, with five more months of play.
Where it is used.
Ports
Salt-accelerated corrosion on cranes, fenders and quay structure, and the condition of moorings.
Energy and substations
Hot connections, vegetation on the perimeter and the condition of insulators.
Water and sanitation
Gates, canals, blocked screens and water level.
Tunnels and viaducts
Water ingress, rockfalls and the condition of joints.
Industrial units
Roofs, downpipes, steel structure and build-up in guttering.
Civil works
Slopes, retaining walls, expansion joints and wall faces.
Ski lifts and gondolas
Anchors, supports and station safety components.
Marine installations
Salt-accelerated corrosion on structure and moorings.
Railways
Containment mesh, overhead line and track components.
The cost of not seeing it in time.
The difference between a planned job and an unplanned one is not the price of the part: it is the shutdown, the emergency call-out and, on critical sites, the service you stop providing.
That is why what matters here is not detecting the break — that already detects itself, when it happens — but gaining weeks of warning. And those weeks are gained by seeing what changes slowly.
We do not promise that everything gets detected. We promise that what can be seen is seen every day, and not every six months.
How you start.
- Three or four critical points, not the whole site.
- One reference image per point, taken in good conditions.
- A month of silent comparison to tune what counts as a change.
- After that, the rest of the points and the report for the inspection.
Fire and smoke, which is another matter entirely.
Fire detection from an image does not replace a fire detection system: that has its own regulations, its certification and its maintenance, and it is compulsory where it is compulsory.
What the camera adds is coverage where there is no detector: outdoors, yards, stockpiles, tall open-plan units where smoke takes a long time to reach the roof. A conventional detector arrives late there, or does not arrive at all.
And it adds the image. A detector says there is smoke; the camera shows what is burning, which changes the response completely.
Vegetation eats the installation.
In substations, canals, tracks and perimeters, vegetation is a maintenance problem on a fixed calendar: it gets cleared twice a year.
But it does not grow the same every year or on every stretch. With the monthly comparison you see where it has really grown and where there was no need to go, and the clearing happens where it is needed.
Drainage and water.
A blocked screen is nothing until the day it rains hard. Then it is a flood and a shutdown.
The blockage shows weeks earlier: leaves, plastic, sediment building up. And the water level in a channel or a bund is read from the image itself without fitting a probe.
With the weather conditions recorded alongside — rain, snow, wind — you also have the context for why what happened, happened.
Casos grabados
Funcionando, en sitios de verdad.
Protege tu infraestructura con la detección inteligente de fugas de Infinity Neural
Infinity Neural: Detección instantánea de fugas para proteger tu infraestructura
Infinity Neural: Detección de fugas con visión por IA para infraestructuras críticas
Detección inteligente de fugas de agua en tiempo real – Infinity Neural
IA - Refuerza el control de válvulas y vertidos con Infinity Neural
Rust Neural: Auditoría avanzada de infraestructuras viales
Control de manómetros y válvulas en estructuras industriales
¡Explora la vanguardia en la gestión de presas con Rust Neural!
Prevención y mantenimiento proactivo de equipos energéticos con Rust Neural
What it does not do.
It does not replace a structural inspection. It does not measure thicknesses or run tests: it sees that something has changed and says when it started changing.
And it does not see what is covered, or the inside of anything. If the critical element is inside a housing, this does not reach it.
And what sits alongside it.
Other things the same family understands, on the same cameras.
What people ask us.
▸Do we have to replace the cameras we already have?
Almost never all of them, and sometimes none. What decides it is not the brand or the megapixels: it is the angle, the distance and the light. A modest camera well placed works better than a good one looking from the wrong spot.
The first thing we ask for is an image from the camera exactly as it is. With that you can see whether it works, whether it has to be moved or whether another one is needed. And if another one is needed, we say so then and not after signing.
▸Where is the video processed?
Wherever it needs to be. It can be on a machine on your own site, with the image never leaving it, or on a server. On sites where the video cannot leave for reasons of policy or regulation, it is processed inside, full stop.
What goes out in that case is not image: it is the alerts and the figures.
▸And data protection?
None of this identifies people. It tells apart figures, postures, objects and vehicles, not faces or names.
Even so, a system that processes images of a workplace has its obligations: informing people, defining what it is used for and how long things are kept. That gets sorted at the start, with whoever handles data protection at the company, and it is not paperwork left to the end.
▸How often is the comparison made?
Daily for what can change fast — blockages, water, rockfall — and weekly or monthly for what advances slowly, like corrosion.
Comparing slow things too often only creates noise: the difference between today and yesterday in a crack is zero, and the system would end up alerting you about shadows.
▸Does it work with a thermal camera?
Yes, and for some things it is the only thing that works. A bearing heating up, an electrical connection with resistance, a patch of lost insulation: none of that shows in a normal image.
Thermal is more expensive and has less resolution, so it goes where it is needed and not everywhere.
▸How does it tell a real change from a shadow?
By comparing images taken in similar conditions and discarding the ones that are not: in fog, in heavy rain or against the light, that reading does not count.
And by requiring persistence: a change that appears once and disappears is not a change, it is a shadow. One that holds for three days is.
▸Is it any use for meeting an inspection plan?
As support, and that alone changes quite a lot: the technician arrives with the series of dated images and knowing which point has changed.
The statutory inspection is still signed by whoever has to sign it, with their measurements and their tests.
▸And fire detection, does it meet the regulations?
No, and it is important to say so. A certified fire detection system is another thing altogether and it is compulsory where it is compulsory.
This covers where a detector does not reach: outdoors, stockpiles, yards and tall units. And it adds the image of what is burning, which a detector does not give you.
▸How quickly does it detect a change?
It depends on how fast the thing changes. A crack advancing millimetres a month shows on the second or third monthly comparison; a rockfall shows the same day.
What it adds is not the speed, it is that the comparison exists at all. Today, between one inspection and the next, there is nothing.
▸Does it count as an official inspection?
No. A structural inspection is signed by an engineer and involves measurements and tests that a camera does not do.
What this does is get the technician there knowing where to look and with a series of dated images, instead of starting from scratch.
▸Does it work at night and in bad weather?
The comparison is made between images taken in similar conditions: midday is compared with midday. In fog or heavy rain that image is discarded and it waits for the next one.
▸How long does it keep the images?
As long as you decide and as long as the drive allows. For this, what is useful is keeping one reference image a day, not the whole video: that is a few megabytes a month per point.
Get started
Do you want this on your site?
This is already recorded working. What is left is to see whether your camera allows it: where it is fitted, what angle it has and what can be seen from there. If it will not work, we will tell you.