2D Item Detection and Sortation
Accurately detect presence or absence, classify, and sort items
Logistics facilities must detect and sort a wide variety of items rapidly and accurately. However, since items can be shipped in a wide range of packaging, including boxes, padded envelopes, and polybags, it’s often difficult to accurately detect presence or absence of an item within a container (tote, tray, etc.), properly classify that item, and correctly sort it to its next destination.
Items can be difficult to detect given the range of packaging types and varying backgrounds of trays or conveyors. But if items are not correctly sorted or routed, they can end up in unexpected areas, such as getting stuck in a cross-belt sorter or conveyor, ultimately causing a jam or equipment damage.
Edge learning can automate the process of detecting, classifying, and sorting packages, so you can avoid these issues.
A subset of AI, edge learning allows processing to take place on-device, or “at the edge,” using a pre-trained set of algorithms. The technology is simple to set up, requiring smaller image sets and shorter training and validation periods than traditional deep learning-based solutions.
Powered by edge learning technology, Cognex's In-Sight 2800 Detector solves difficult 2D item detection and object sortation applications with simple example-based training and powerful AI algorithms.
This system can accurately detect item presence or absence in ship sorters or processes using totes or trays, even with varying backgrounds or low contrast. Edge learning allows the In-Sight 2800 Detector to be trained on any number of packaging types to enable proper sortation in inbound and outbound processes. Process issues can also be detected such as presence of items in the bellow of a cross-belt sorter or a conveyor jam.