How Many Cameras Does Your Inspection Machine Need?
Buyers often ask for a camera count. The honest answer starts from the defect, not the number.
Camera quantity depends on three things: the inspection area, the defect type and the inspection tolerance. A cap top alone is a different problem from a cap top plus sidewall plus internal thread, and the second needs more views not because it is a better machine but because a camera cannot see around a corner.
Camera resolution is chosen from your precision requirement and the product itself. A 0.2 mm by 0.2 mm defect and a 0.5 mm by 0.5 mm defect do not need the same sensor. We select the definition, for example two megapixel or three megapixel, according to the inspection request rather than fitting the highest resolution available and charging for it.
Most of our cameras are designed and developed in house, because when the inspection system is AI based the camera has to match the AI system. Lenses and sensors come from worldwide brands. We remain responsible for the technical support and after-sales service of the hardware either way.
The camera is one of the most important hardware components, but the AI algorithm software is the key technology of the machine. A high camera count with a weak algorithm produces a high false reject rate, which operators solve by loosening the standard until the machine stops helping.
More technology notes
- Good Product Training: How AI 2.0 Inspection Handles Defects You Have Never Seen
- Intelligent Defect Generation: Three Steps to a Training Set You Do Not Have
- AI Algorithms Versus Traditional Algorithms in Visual Inspection
- How a Visual Inspection System Is Customised in Four Testing Stages
- What Defects a Vision Inspection Machine Can Actually Detect
- Inline or Offline Inspection: Which One Fits Your Plant
- What We Need From You Before We Can Quote
Apply this to your line
Send your parts and defect samples and we will tell you which approach fits, and what it will and will not catch.