click here to download project abstract
– ABSTRACT
Plant species detection aims at the automatic identification of plants. Although a lot of
aspects like leaf, flowers, fruits, seeds could contribute to the decision, but leaf
features are the most significant. As a plant leaf is always more accessible as
compared to other parts of the plants, it is obvious to study it for plant identification.
The present paper introduced a novel plant species classifier based on the extraction
of morphological features .First, a two-layer plant taxonomy is constructed to organize
large numbers of plant species and their genus hierarchically in a coarseto-fine
fashion. Second, a deep learning framework is developed to enable path-based tree
classifier training, where a tree classifier over the plant taxonomy is used to replace
the flat softmax layer in traditional deep CNNs. A path-based error function is defined
to optimize the joint process for learning deep CNN and tree classifier, where back
propagation is used to update both the classifier parameters and the network weights
simultaneously.
