{"id":3105,"date":"2022-07-13T09:30:00","date_gmt":"2022-07-13T07:30:00","guid":{"rendered":"https:\/\/cirpicme.org\/?page_id=3105"},"modified":"2022-07-26T13:04:27","modified_gmt":"2022-07-26T11:04:27","slug":"interactive-image-segmentation-using-superpixels-and-deep-metric-learning-for-tool-condition-monitoring","status":"publish","type":"page","link":"https:\/\/cirpicme.org\/index.php\/cutting-technologies-2\/interactive-image-segmentation-using-superpixels-and-deep-metric-learning-for-tool-condition-monitoring\/","title":{"rendered":"Interactive Image Segmentation Using Superpixels and Deep Metric Learning for Tool Condition Monitoring"},"content":{"rendered":"\n

by Benjamin Lutz, Lucas Janisch, Dominik Kisskalt, Daniel Regulin, Joerg Franke (Germany)<\/em><\/p>\n\n\n\n

<\/p>\n\n\n\n

Abstract<\/strong><\/p>\n\n\n\n

The optical measurement of tool wear is commonly used to monitor machining processes. Recently, deep learning methods, in particular, have been applied for the identification and segmentation of the different wear defects. For such approaches, annotated training data is required, which includes the acquisition of cutting tool images and their pixel-wise annotations. As this process is time-consuming, we propose a novel interactive image annotation method for tool condition monitoring. This is achieved by combining superpixel segmentation, deep metric learning, and human corrections. By using the interactive image annotation, manual annotation effort is reduced and mask accuracy is improved.<\/p>\n\n\n\n

Keywords<\/strong>: Interactive Segmentation; Superpixel Segmentation; Deep Metric Learning; Tool Condition Monitoring; Smart Manufacturing; Human-in-the-loop<\/p>\n\n\n\n

<\/div>\n\n\n\n

Video presentation<\/strong><\/p>\n\n\n\n