Recognition of 3D tree crown structure based on its pictures: statistical approach
Alexei A. Ivankov, Mikhail A. Kurochkin · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1998
The investigation is devoted to the recognition of 3D object structure on the basis of the object pictures. Recognition problem is solved as statistical one, since the nature of the structures are stochastic most likely. In its turn the images are damaged with noise. Any structure under recognition is assumed to be a trajectory of a stochastic markovian process. A priori necessary information is the set of the values of the markovian process parameters to determine uniquely the general population of all structures for current recognition session. The significance level as input parameter serves to recognize certain domain of the general population the recognized structure matches with prescribed probability. Our method (algorithm and software) could gain wide-spread acceptance in object visual control, ecology monitoring, as well as to resolve the problem of some alive object growth description, i.e. plants and trees growth.