Based on deep learning and image segmentation algorithm, defect features are identified and located in real time, and OK/NG judgment results are output in real time. -Zero foundation, easy to use Quickly create models through an intuitive UI interface -Fast training model The model can be trained with only 5 samples for each defect type -Adjustable threshold parameters Accurate detection of pixel-level defect and adjustment of threshold parameter can be realized -Scale across plant level It can realize the large-scale deployment of AI model in the global production line.
English (UK)
