A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting.
Lonesome MalamboSorin C. PopescuNian-Wei KuWilliam L. RooneyTan ZhouSamuel MoorePublished in: Remote. Sens. (2019)
Keyphrases
- deep learning
- semantic segmentation
- weakly supervised
- superpixels
- unsupervised learning
- conditional random fields
- machine learning
- object class
- topic models
- object categories
- relation extraction
- decision trees
- semi supervised
- active learning
- pairwise
- named entities
- semantic relations
- object recognition
- multiscale
- training data