Machine vision for laser defect in PV solar modules

This paper presents a new methodology in inspection on laser scribe defect of PV thin film solar modules. The work focuses on the application of machine vision as an inspection tools which has successfully integrated in other manufacturing environment as pattern recognition utility. Compared to manual inspection by human, machine vision system could offer better measurement accuracy as scribe defects are extremely hard to detect due to their small sizes and complexity of the detection process. Studies were made to identify machine vision system screening capabilities to define different scribe defect by their inspection criteria. Current result with paper and broad samples indicates that the propose system can be used effectively to replace human evaluators that currently employs in manufacturing quality control.