Cameras, optics, lighting, and triggering engineered as one system
Most industrial camera projects fail on integration, not imaging: the right camera behind the wrong lens, frames dropped under load, triggers that drift, cables that die from flexing. Camera system integration is the discipline of making the whole acquisition chain — sensor, optics, lighting, trigger, transport, and mounting — work as one reliable instrument.
Willowark integrates cameras into machines, inspection stations, vehicles, and instruments. We do the unglamorous engineering first: resolution and working-distance math, interface bandwidth budgets, trigger timing diagrams, and thermal and vibration analysis, so the system you commission is the system that still works in month eighteen.
Vision & Advanced SensingHow the work gets done
The same way every time: scope, build, hand over.
Selection starts with sensor format and pixel pitch matched to the lens image circle and mount — a mismatched pair wastes resolution you paid for. Interface choice follows the bandwidth budget: GigE Vision with PoE for long cable runs and multi-camera networks, USB3 Vision for short-haul simplicity, CoaXPress when frame rate and resolution outrun both. Hardware triggering from encoders or photo-eyes, strobe timing matched to exposure, and PTP (IEEE 1588) synchronization keep multi-camera captures aligned. On the PC side, NIC tuning, jumbo frames, and receive buffers are configured so frames survive real system load, not just bench tests.
Deployment is where camera systems earn their keep. We specify IP-rated housings and lens tubes for washdown or dusty environments, vibration-isolated mounts that hold calibration, industrial cordsets with proper strain relief, and acquisition watchdogs that detect and recover a dropped camera without an operator noticing. Handover includes cabling diagrams, spares strategy, and calibration procedures your maintenance team can actually follow.
We prove the acquisition chain on the bench before it goes on the machine. A short imaging trial with your parts, candidate lenses, and lighting answers the questions that datasheets cannot: whether the depth of field covers your part height variation at the aperture the exposure time needs, whether the strobe is bright enough to freeze motion at your line speed, and whether the chosen sensor has the dynamic range to hold both the dark and bright regions of the part. Lens, aperture, exposure, and light intensity trade against each other, and the trial finds the combination with margin rather than the one that barely works.
The software side of integration is mostly about not losing frames and knowing when you have. Acquisition code is built on the vendor SDK or GenICam and GenTL so cameras can be swapped without a rewrite, and every frame carries a hardware timestamp and frame counter so gaps are detected and logged rather than silently skipped. We soak-test the finished chain for extended runs at full rate under realistic CPU load, and record dropped-frame counts, temperatures, and trigger latency as the acceptance evidence. Handover includes the source code, a documented camera configuration file, and a step-by-step procedure for replacing a camera or lens and re-verifying focus and alignment.
Scope it in writing
What we agree before work starts
- Imaging requirements analysis with optics and resolution calculations
- Camera, lens, lighting, and interface specification with bandwidth budget
Build with checkpoints
Working results, not slide decks
- Trigger and synchronization design with timing documentation
- Mechanical mounting and enclosure design for the environment
Hand over something you own
Documentation, source, and training
- Acquisition software or SDK integration with recovery watchdogs
- Installation, commissioning, and maintenance documentation
Sound familiar?
Where camera system integration earns its keep.
Adding inspection cameras to an OEM machine design
Multi-camera 360-degree inspection of parts in a single station
High-speed capture of intermittent events for process diagnosis
Retrofitting aging analog or obsolete cameras with modern digital systems
Ask about Camera System Integration
Describe the problem. Get a straight answer.
One line is enough. An engineer replies within a business day.
Related work
Radar centering system for steel mills
Components:
- Radar sensors (strip position): Non-contact radar reading strip edge position in a hot, dusty, vibrating environment where optical sensors fail.
- Edge controller (signal processing): Turns raw radar returns into a clean lateral offset in real time.
- Mill PLC (centering actuators): The mill's existing controller: receives the offset and drives the centering actuators.
- Operator HMI (live position): Live strip position for the operator.
Connections:
- Radar sensors to Edge controller (raw returns)
- Edge controller to Mill PLC (offset)
- Edge controller to Operator HMI
A steel-mill systems provider · Steel manufacturing
Radar-based centering system for steel mills
End-to-end engineering of a radar sensing system that measures and centers material on steel mill lines — from equipment assessment through hardware selection, electrical engineering, software, installation, and commissioning.
Read the case study →Common questions
Asked before every camera system integration project.
GigE or USB3 — which should we use?
It comes down to cable length, bandwidth, and camera count. GigE Vision runs 100 meters over standard Ethernet, powers cameras via PoE, and scales to many cameras on one network. USB3 offers more bandwidth per camera but is practically limited to a few meters. We run the numbers for your frame rate and layout before recommending either.
Can you integrate cameras with our existing PLC or software?
Yes. We regularly tie acquisition into PLCs over EtherNet/IP or PROFINET for triggering and results, and integrate with existing HMI, MES, or custom software through the camera vendor SDK or standard protocols. The camera system becomes another well-behaved device on your machine, not an island.
Our environment is hot, dusty, and gets washed down. Is that a problem?
It is a design input, not a blocker. IP67 housings, air-purged lens ports, thermal management, and sealed connectors are standard tools for harsh environments. The key is specifying them up front — retrofitting environmental protection after cameras start failing costs far more.
Do you work with a specific camera brand?
No. We select from the established industrial camera and lens manufacturers based on sensor, interface, availability, and your existing spares, and we build acquisition code on GenICam-standard interfaces so a future camera swap is a configuration change rather than a rewrite. If your plant already standardizes on one brand, we typically work within that standard unless it genuinely cannot meet the imaging requirement.
How do we keep the cameras calibrated after installation?
Mechanically first: rigid mounts, locked lens rings, and reference marks so a bumped camera is obvious. Then procedurally: a reference target or part imaged at a set interval, with the software comparing focus, position, and brightness against the commissioned baseline and flagging drift before it affects results. The handover documentation includes the re-alignment procedure, and for measurement applications, the calibration artifact and the steps to re-establish scale.
Where this sits
Camera System Integration, inside a vision & advanced sensing system.
The lit component is the part of the system this service delivers; the rest is what it has to work with.
Hover or focus a component to see what it is and what it talks to. Arrow keys move between them.
A part is presented under controlled lighting, a camera captures a frame per trigger, inspection compute decides, the PLC rejects, and every result lands in a database that feeds SPC dashboards.
Components:
- Part (on conveyor): Presentation is half the problem: fixturing, orientation and cycle time decide what is possible.
- Lighting (ring / backlight): Chosen for the defect, not the camera. Lighting is where most vision projects are won or lost.
- Camera (GigE, triggered): Machine vision camera, hardware-triggered per part.
- Inspection compute (edge PC): Runs the inspection — classical tools, a trained model, or both — within cycle time.
- Line PLC (reject / accept): Acts on the verdict: reject gate, line stop, or count.
- Results DB (every part): Every inspection result, with the image reference, for traceability and SPC.
- SPC dashboard (trends): Escape rate, false-reject rate and drift over time.
Connections:
- Part to Camera over digital I/O (trigger)
- Lighting to Camera
- Camera to Inspection compute over GigE
- Inspection compute to Line PLC over digital I/O
- Inspection compute to Results DB over SQL
- Results DB to SPC dashboard over REST
Strategy. Software. Systems.
Have a system that should exist?
Tell us what your operation is doing manually, what isn't connected, or what you're trying to build. We'll tell you plainly whether and how we can help.