Abstract
The study discusses adaptive neuro-fuzzy methods of recognition of the multidimensional overlapping objects on the basis of the introduced concepts, generalized and modified operations of fuzzy set theory and neural networks. To improve recognition accuracy, proposed a combined approach including neural network analysis of generalized images based on Kohonen maps, and building systems fuzzy inference based on the identification of allocated clusters integral characteristics of the images. Using the derived system of diagnostic decision rules "If ... then" the comprehensive forecast map of oil South of Cheremshan field was built.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have