Open topic with navigation
ROCK MAKER allows you to score your images automatically with MAchine Recognition of Crystallization Outcomes (MARCO). MARCO is a machine learning algorithm that can classify images as one of two image classes: Crystal and Non Crystal. MARCO was developed by a team of researchers from Google Inc., in collaboration with academia and industry specialists in the US and Australia.1Bruno AE, Charbonneau P, Newman J, Snell EH, So DR, Vanhoucke V, et al. (2018) Classification of crystallization outcomes using deep convolutional neural networks. PLoS ONE 13(6): e0198883. https://doi.org/10.1371/journal.pone.0198883 You can configure the MARCO scoring settings in the Well Score Configuration table of the General Setup.
MARCO can only score images that have been imaged using Visible light. You can send all of the visible images from an inspection for scoring with MARCO at once. If your ROCK MAKER software is integrated with ROCK IMAGER, all visible images in an inspection will be automatically sent for MARCO scoring. See Setting Up MARCO and System Properties for more information. Please contact email@example.com if you need additional help setting up MARCO.
When MARCO is enabled, the Send to MARCO button will automatically appear in the Scores tab of the Auto Score group in the Images contextual tab set. The Dispute and Send for Training buttons will also appear if you wish to help improve MARCO. To participate in improving MARCO, you must go to the General Setup, and select the Send images for MARCO training and improvement. option on the Feedback for Improvement table at the very bottom of the General page. Read Sending Images for MARCO Training for more information.
Send to MARCO, Dispute, and Send for Training Buttons in Scores Tab of the Images Tab Set
1Bruno AE, Charbonneau P, Newman J, Snell EH, So DR, Vanhoucke V, et al. (2018) Classification of crystallization outcomes using deep convolutional neural networks. PLoS ONE 13(6): e0198883. https://doi.org/10.1371/journal.pone.0198883