Study on Plant Leaf Recognition Method Based on Multi-feature Fusion

Aiming at the problem that a single feature cannot accurately describe plant leaves, a plant leaf recognition method based on multi-feature fusion is proposed in this paper. In view of the local binary pattern feature extraction method and two improved algorithms, considering implications of multi-neighborhood pixels toward middle pixel, a new mixed weighted neighboring local binary pattern texture descriptor is constructed, and the histogram of oriented gradient feature is fused to gain a more detailed leaf feature description vector. Finally, the experimental results on the open Flavia data set show that the proposed method can obtain good experimental results.