CLASSIFYING PRODUCT BY SIZE AND COLOR
Solve sorting challenges quickly and efficiently
Graphical programming environment for deep learning-based industrial image analysis
During the manufacturing process, consumer goods may require sorting by size and color for kitting or case packaging. For example, a soap bar manufacturer produces several scents (variants) of the same brand of soap. These products may be packaged homogenously in cases or may be packaged as an assortment. Variables like lighting, background color, and size can affect the way an object appears to the camera, camouflaging variations and challenging vision system’s identification capabilities. Similar-looking parts of the same size but slightly different colors can cause picking and packaging errors, which can lead to customer dissatisfaction, loss of brand reputation, and skew inventory levels. Manual sortation is challenging and unreliable given the high speed at which consumer products are produced.
Using edge learning technology to solve sorting challenges is a quick and efficient way to automate the sorting process. The intuitive classification tool is trained with images of different colored packages so once in production, the system quickly identifies the color of each product and routes it accordingly. The edge learning-based sorting system minimizes changeover time as the system can quickly be retrained using only a few new images of the upcoming product. Automating the sorting process decreases packaging errors, helps ensure end-product quality, and reduces re-work costs.