论文信息 - Corn Disease Identification Based on improved GBDT Method

Corn Disease Identification Based on improved GBDT Method

Leaves images of diseased and healthy plants are play an important role in the plant disease detection and diagnosis. This study proposed a Gradient Boosting Decision Tree (GBDT) method for corn disease identification. Firstly, we use Synthetic Minority Over-Sampling Technique (SMOST) method to solve the problem of data imbalance. Then, we select regional interpolation algorithm to resize the leaf image resolution to 30×20. Finally, gradient boosting decision tree algorithm is employed to corn disease identification. The experiment results on Northern Leaf Blight-Infected dataset indicated that the proposed method is effective, and achieves 92.5% accuracy.

[1] Alex Alves Freitas,et al. An empirical evaluation of hierarchical feature selection methods for classification in bioinformatics datasets with gene ontology-based features , 2018, Artificial Intelligence Review.

[2] Anup Vibhute,et al. Applications of Image Processing in Agriculture: A Survey , 2012 .

[3] Mohamed Elhoseny,et al. Feature selection based on artificial bee colony and gradient boosting decision tree , 2019, Appl. Soft Comput..

[4] J. Yosinski,et al. Automated Identification of Northern Leaf Blight-Infected Maize Plants from Field Imagery Using Deep Learning. , 2017, Phytopathology.

[5] Tie-Yan Liu,et al. LightGBM: A Highly Efficient Gradient Boosting Decision Tree , 2017, NIPS.