Finding a paint defect is only half the job. A crater, a sag and a dirt nib can sit side by side on one panel, yet each was born somewhere different in the booth and each needs a different repair. If the inspection line only says "defect", the cause keeps producing more of them and the repair deck guesses at the fix. Paint defect detection with AI vision names the defect, points to its likely root cause in the booth and sends the body to the right repair. See it on real panels in a short paint vision demo.
Name the Defect, Find Its Cause in the Booth, Route the Repair
iFactory Paint Vision AI uses tunnel cameras to classify every defect, checks it against booth humidity and temperature and routes each body to the right repair by defect type.
Six Defects That Fill Most Repair Decks
Most paint shop rework comes from a short list of defects. Each one has a typical look and a typical birthplace, which is why classifying them correctly is the first step to fixing the process, not just the panel. Coatings makers list causes such as excess film build and low temperature for runs, high temperature and high viscosity for orange peel, and silicone or oil contamination for craters. Review your own top defects with our paint quality team.
From Defect Type to Likely Root Cause
Classification becomes useful when every defect class is linked to the process data that can confirm or rule out a cause. iFactory stores booth, robot and material data against each body, so the likely cause comes with evidence instead of a guess. The matrix shows typical links. Confirm them on your own line in a root cause session.
How Booth Conditions Shift the Defect Mix
Booth temperature and humidity change how fast solvents and water evaporate, and that changes which defects appear. Cold, damp air slows evaporation, so paint stays wet longer and is more likely to sag. Hot, dry air speeds it up, so paint has less time to level and orange peel and solvent pop become more likely. The grid shows general tendencies, not limits, since every paint system has its own window. Find your own window with our process specialists.
Where the Dirt Comes From, Read From Where It Lands
Dirt and inclusions are often among the most frequent defects on a paint line, and the hardest to blame on one source. Position on the body is a strong clue. Fibres on horizontal surfaces suggest something falling from above, nibs low on the sides point to the conveyor or floor, and dirt in one zone on every body points to one robot or one booth section. iFactory maps every nib by panel and zone, so patterns show up within a shift. Trace your dirt sources in a dirt mapping session.
Light Tunnels and Deflectometry, Side by Side
Classic inspection uses light tunnels, where trained inspectors look for defects under bright stripes of light. Deflectometry automates the same idea: a known stripe pattern is reflected off the body and cameras measure how the reflection bends, which reveals small bumps and dents in the surface. iFactory tunnel cameras feed classification models that label each defect and record its position on the body. Plan your tunnel setup with our vision engineers.
- Inspectors judge reflected light by eye
- Results depend on skill, fatigue and line speed
- Defects marked by hand, often without a record
- Bent stripes show bumps and dents as data
- Every body checked the same way
- Each defect logged with class, size and position
Every Defect Arrives With a Cause and a Route
iFactory tunnel cameras classify each defect, correlate it with booth humidity and temperature at the time the body was painted and route the body to the right repair by defect type. Booth leads see the pattern while it is still forming.
Dry Film Thickness: The Number Behind Runs and Pops
Too much film build drives runs, sags and solvent pop, and too little causes poor hiding and weak protection. Dry film thickness is usually measured with magnetic or eddy current gauges, as described in ASTM D7091, and ISO 2808 sets out the wider range of methods. Wet film readings help catch problems before the oven, because dry film thickness is roughly wet film thickness multiplied by the paint's volume solids. Link your gauges to iFactory in a data connection review.
Sending Each Body to the Right Repair
Not every defect needs the same fix. A dirt nib in the clearcoat can often be sanded and polished in minutes, while a crater that goes through the colour layer needs a spot repair, and a large sag on a visible panel may need a full panel repaint. iFactory routes each body using defect type, size, layer and zone, following the repair rules your quality team sets. Map your repair rules with our quality specialists.
A Defect Record That Closes the Loop
Every defect record should connect the finished panel to the booth that painted it and the repair that fixed it. Use these fields as a template, with or without AI. Ask our paint vision team for a blank version.
Found, Traced, Routed: Three Illustrations
These examples show how the system typically responds. They are illustrations, not case studies. Walk through your own recent defects with our paint engineers.
A Worksheet for Plant Managers
The value comes from two places: fewer defects once causes are fixed, and cheaper repairs when bodies go straight to the right route. Every input below is assumed, so replace each one with your own figures. Build the full case with our ROI specialists.
Multiply over-repairs by the cost difference between routes, then add the value of each root cause fixed. Keep the sum to figures your own repair data supports.
Tunnel Cameras, Booth Links and NVIDIA Server in One Package
iFactory delivers Paint Vision AI as a turnkey bundle. Tunnel cameras and lighting are installed after the oven, booth sensors are linked, and a pre-configured NVIDIA AI server ships racked and ready with the defect and root cause models pre-loaded, so every body is classified on site. You rack it, plug in power and Ethernet, and the AI is live. To scope it for your paint shop, request a turnkey quote.
- Pre-configured NVIDIA AI edge server, racked and ready
- Tunnel cameras and lighting for painted bodies
- Defect classification models trained on your finishes
- Booth temperature and humidity correlation
- Film build and gauge data links
- Repair routing rules by defect type and zone
- 24×7 remote monitoring of models and hardware
Tunnel cameras and server installed, booth data linked, images collected.
Defect models trained and checked against inspector calls on the same bodies.
Repair routing and booth alerts switched on, teams trained.
Paint Defect Detection Software That Fixes the Booth Too
iFactory links what the camera sees on the panel to what happened in the booth. Our support team can walk you through a pilot plan.
What Plant Managers Ask About Paint Defect Detection
Fix the Panel and the Booth That Painted It
Classify every paint defect, link it to booth humidity and temperature and send each body to the right repair. The full system is delivered turnkey and goes live in 6 to 12 weeks.
