A flood mapping algorithm from cloud contaminated MODIS time-series data using a Markov random field model.
Teerasit KasetkasemPonlapak PhuhinkongPreesan RakwatinThitiporn ChanwimaluangItsuo KumazawaPublished in: IGARSS (2014)
Keyphrases
- preprocessing
- experimental evaluation
- computational cost
- learning algorithm
- k means
- optimal solution
- dynamic programming
- theoretical analysis
- computationally efficient
- optimization algorithm
- times faster
- objective function
- data sets
- np hard
- noisy data
- simulated annealing
- input data
- expectation maximization
- improved algorithm
- segmentation algorithm
- remote sensing
- mapping function
- clustering method
- detection algorithm
- graph cuts
- worst case
- scheduling problem
- hidden markov models
- cost function
- multi objective
- computational complexity
- reinforcement learning