Model selection and timing of acquisition date impacts classification accuracy: A case study using hyperspectral imaging to detect white pine blister rust over time.
Marja HaagsmaGerald F. M. PageJeremy S. JohnsonChristopher StillKristen M. WaringRichard A. SniezkoJohn S. SelkerPublished in: Comput. Electron. Agric. (2021)
Keyphrases
- model selection
- hyperspectral imaging
- classification accuracy
- target detection
- feature selection
- cross validation
- machine learning
- remote sensing
- sample size
- hyperspectral
- motion segmentation
- selection criterion
- training set
- model selection criteria
- support vector
- infrared
- hyperspectral images
- detection method
- information criterion
- training data
- feature subset
- data sets
- feature space
- false alarms
- hyperspectral data
- bayesian information criterion
- marginal likelihood
- image analysis