TRAINING AN ARTIFICIAL NEURAL NETWORK USING SIMULATED SPECIMEN IMAGES

摘要

Disclosed is a technique for training an artificial neural network (ANN) by using a simulated sample image. The simulated sample image is generated based on a data model. The data model describes properties of a crystalline material and properties of at least one defect type. The data model does not contain any image data. The simulated sample image is inputted to a training algorithm as training data to create an ANN for identifying a defect in the crystalline material. After the ANN is trained, the ANN analyzes the inputted sample image to identify a defect in the image.