Accurately counts filters in shipping boxes no matter the filter size, color or shape
Graphical programming environment for deep learning-based industrial image analysis
Tobacco manufacturers receive cigarette filters in bulk in large boxes from filter makers. To receive shipments into stock, they measure the box sizes to get an approximate count of the number of filters inside. However, this technique is not accurate and can result in over- or undercounts.
Cognex Deep Learning solves this challenging application by counting 100% of filters in large shipping boxes. Combining the location tool with a 29MP camera mounted overhead and powerful external bar lights positioned parallel to the box, Cognex counts boxes containing over 4,000 filters. Cognex Deep Learning is easily trained using a small sample set of sample images to identify defects on all types of filters, including white, charcoal and recessed, to ensure accurate counts no matter the filter size, color or shape.