Keyence has added the IV4-H to its IV4 series of all-in-one vision sensors, building artificial intelligence directly into the device to automate the parameter tuning that machine vision engineers have historically handled through manual trial and error. Keyence says the sensor evaluates more than 10,000 combinations of exposure, aperture, and image-filter settings to find the clearest picture for a given inspection, a process that can otherwise consume hours of setup time.
The IV4-H also uses AI to identify the specific area it needs to inspect and filter out background clutter, a common cause of false positive and false negative calls in conventional vision systems. That addresses one of the more persistent problems on the shop floor: engineers spend time troubleshooting nuisance rejects and missed defects instead of running inspections. Keyence says the noise filtering also helps stabilize results in harsh environments where lighting and vibration vary.
Like other sensors in the IV4 line, the IV4-H integrates its own lighting and applies AI across every function, whether it is identifying parts, flagging flaws, reporting positional data, reading text and numbers, or counting objects. Keyence increased the brightness of the built-in illumination for this model and added flexible lens options to cover a wider range of fields of view.
Setup gets a further shortcut through an AI trigger that detects an incoming part and fires the camera at the optimal moment, removing the need for a hardwired or control-system trigger signal. The sensor head runs on Power over Ethernet, handling power and data over a single cable, and carries an IP67 rating. Keyence also offers optional high-flex cables for installations on robotic arms.
Keyence points to assembly verification as one use case for the IV4-H, where the sensor checks that an assembly contains the correct set of components rather than just flagging a single surface flaw.
