Fault detection and localization in empty water bottles through machine vision

The paper presents a vision-based approach for fault detection of empty water bottles. Our goal is to find all defects in the empty water bottles. The main constraint is the real time operation as the bottles move continuously on the conveyer belt. Using camera images, we first find the position of the bottle in the image by a generalized Hough transform. Then we check the bottles for any defects or the presence of alien objects using rectangular and circular area intensity inspection. Bottles with any defects are rejected and so separated from the main stream. In the statistics obtained, our inspection system successfully rejected all the faulty or defective bottles. The good bottle rejection rate was lower than 1% of the total number of bottles inspected. This system can be used for inspecting any type of empty bottles for defects or the presence of alien objects.