Using “Just Noticeable Difference” to Automate Visual Inspection of Displays According to Human Visual Perception


Human perception is the ultimate standard for determining the visual quality of a device. However, the use of human inspection as a quality control method for development or production of devices is problematic because of the statistical variation between observers. In this article, learn how human visual perception guides display testing, which is rapidly being automated using photometric and colorimetric imaging systems capable of objectively quantifying visual qualities like brightness, color, and contrast of displays. In particular, these systems employ standardized human-centric measurement methods such as "Just Noticeable Difference" (JND) analyses to assign values to areas of unnevenness in displays that may be just noticeable to a standard human observer, enabling automated pass/fail operations for display quality control.

Published by Quality Magazine on January 8, 2020.