Self-learning adaptive algorithm for maritime traffic abnormal movement detection based on virtual pheromone method.
Julius VenskusMindaugas KurmisArunas AndziulisZydrunas LukosiusMiroslav VoznákDenisas BykovasPublished in: SPECTS@SummerSim (2015)
Keyphrases
- detection method
- detection algorithm
- high accuracy
- cost function
- theoretical analysis
- preprocessing
- clustering method
- improved algorithm
- significant improvement
- estimation algorithm
- segmentation method
- dynamic programming
- experimental evaluation
- similarity measure
- computational cost
- synthetic and real images
- computational complexity
- recognition algorithm
- classification method
- segmentation algorithm
- optimization algorithm
- objective function
- k means
- energy function
- computationally efficient
- line detection
- learning algorithm
- tree structure
- anti noise
- reconstruction method
- em algorithm
- mathematical model
- matching algorithm
- convergence rate
- selection algorithm
- input data
- probabilistic model
- classification algorithm
- optimization method
- support vector machine svm
- anomaly detection
- np hard
- search space
- particle swarm optimization
- evolutionary algorithm
- adaptive threshold
- clustering algorithm