A neural network classifier based on prior evolution and iterative approximation used for leaf recognition

The intelligent recognition of leaves is a fast and effective plant classification method. However, the existing classifiers are not completely applicable. On the one hand, the input forms of leaf features are complex; on the other hand, leaf recognition is rather difficult, and the classifier need to provide a list of results, which are arranged based on their possibilities in descending order, for users´ selection, so that the credibility of the result increases. Herein, we propose a new classifier that is a neural network classifier based on prior evolution and iterative approximation, which can satisfy the aforementioned special requirements. Through the analyses of experiments, it is proved to do well in leaf recognition and shows its good classification performance.