Dual Scale Automaton Model for Virtual Plant Development

A variety of plant growth models or methods for plant image synthesis have been reported. However, most of them are based on computer graphics to demonstrate natural scenery, and they are not suitable for the study of agronomy or botany. Although L\|system and reference axis technique can be used to model plant growth, these two models are not easy for understanding and applications. In this work, a dual\|scale automaton model for virtual plant growth is developed from the viewpoints of botany. Two concepts, namely, microstate and macrostate, are proposed according to the physiological age of plant growth. The macrostate corresponds to growth unit and the microstate to metamer. The transitions between or within the microstate and macrostate are represented in semi\|Markov chain. A probability process, in accord with the growth of plant's apical bud and axillary bud, is applied in this model. Using this methodology, the topological structures at each growth stage can be generated. The parameters of this model have explicitly physical meaning for relating to the topological structures and can be given with the help of diagram. All of these features make it easy for users to understand and implement in applications. This model has been tested for all plant architectural models defined by botanists. Only one example is given in the paper to confirm the advantages of the methodology. The present model demonstrates a simpler but more effective method in comparison with other approaches, such as L\|system and reference axis technique.