Glossary · Machine vision
Frame rate
Also known as: Frames per second (fps), Image rate
German: Bildrate
In imaging, frame rate is the number of images a camera can capture and transmit per second, usually given in frames per second (fps). It depends on sensor readout, exposure time, region of interest and interface bandwidth.
- Machine vision
In one sentence
Frame rate is the number of images a camera captures and delivers per second, limited by sensor readout, exposure and bandwidth.
Example
Reducing the region of interest to the label area raised the frame rate from 60 to 220 fps, enough for the faster packaging line.
How it applies
- Engineering: The required frame rate follows from parts per second and images per part; processing time must also fit, otherwise images are dropped. Network cameras share bandwidth with other traffic.
- Commissioning: Triggered acquisition is usually preferred over free-running frame rate in inspection, so each image matches one part.
- Documentation: The documentation team should record the required and achieved frame rate, processing time and the reaction when an image is lost, since dropped images can let parts pass unchecked.
Frame rate vs. latency
Frame rate is how many images per second; Latency is the delay from trigger to result. A camera can have a high frame rate and still deliver results too late for a reject gate.