J.C. Briceño, C. Travieso, and M.A. Ferrer (Spain)
HMM Contour Leaves Image Processing, Size Probabilities
In this paper we present a simulation of a recognition process with perimeter characterization of simple plant leaves as a unique discriminating parameter. An HMM (Hidden Markov Model), performs the main recognition processes. Data coding allowing for independence of leaves size and orientation, may penalize performance recognition for some varieties. Border description sequences are then complemented with probabilistic sizes descriptions. Finally, decision making is established by a weighed mean function on HMM and size probabilities, obtaining a performance ratio of a 78%
Important Links:
Go Back
