Is Your Training Data Really Ground Truth? A Quality Assessment of Manual Annotation for Individual Tree Crown Delineation.
Janik SteierMona GoebelDorota IwaszczukPublished in: Remote. Sens. (2024)
Keyphrases
- quality assessment
- tree crown
- manual annotation
- ground truth
- manually annotated
- aerial images
- training data
- power line
- fully automatic
- image quality
- error prone
- automatic annotation
- automatic analysis
- human visual system
- semi automatic
- automatic image annotation
- data quality
- learning algorithm
- supervised learning
- image annotation
- domain knowledge
- training set
- training examples
- labeled data
- high quality
- semi supervised
- image data
- feature extraction
- image segmentation
- feature selection
- data sets