Saw-Mark Defect Detection in Heterogeneous Solar Wafer Images using GAN-based Training Samples Generation and CNN Classification.
Du-Ming TsaiMorris S. K. FanYi-Quan HuangWei-Yao ChiuPublished in: VISIGRAPP (5: VISAPP) (2019)
Keyphrases
- training samples
- representative samples
- defect detection
- training set
- feature space
- supervised learning
- decision boundary
- image classification
- training data
- nearest neighbor classifier
- high dimensional
- learning algorithm
- test sample
- input image
- sample set
- number of training samples
- feature extraction
- discriminative information
- image features
- hyperplane
- test images
- support vector machine svm
- training examples
- small number
- decision trees
- keypoints
- small number of samples
- support vectors
- active learning
- machine learning
- similarity measure
- reinforcement learning
- image sequences
- decision function
- feature vectors
- classification accuracy
- support vector machine
- multi class
- high dimensionality
- data sets