Abstract:The construction of a pest detection algorithm is an important step to couple “ground-space” characteristics, which is also the basis for rapid and accurate monitoring and detection of pest damage. In four experimental areas in Sanming City, Jiangle County, Sha County and Yanping District in Fujian Province, sample data on pest damage in 182 sets of Dendrolimus punctatus were collected. The data were randomly divided into a training set and testing set, and five duplicate tests and one eliminating-indicator test were done. Based on the characterization analysis of the host for D. punctatus damage, seven characteristic indicators of ground and remote sensing including leaf area index, standard error of leaf area index (SEL) of pine forest, normalized difference vegetation index (NDVI), wetness from tasseled cap transformation (WET), green band ( B 2 ), red band ( B 3 ), near-infrared band ( B 4 ) of remote sensing image are obtained to construct BP neural networks and random forest models of pest levels. The detection results of these two algorithms were comprehensively compared from the aspects of detection precision, kappa coefficient, receiver operating characteristic curve, and a paired t test. The results showed that the seven indicators all were responsive to pest damage, and NDVI was relatively weak; the average pest damage detection precision of six tests by BP neural networks was 77.29%, the kappa coefficient was 0.6869 and after the RF algorithm, the respective values were 79.30% and 0.7151, showing that the latter is more optimized, but there was no significant difference ( p > 0.05); the detection precision, kappa coefficient and AUC of the RF algorithm was higher than the BP neural networks for three pest levels (no damage, moderate damage and severe damage). The detection precision and AUC of BP neural networks were a little higher for mild damage, but the difference was not significant ( p > 0.05) except for the kappa coefficient for the no damage level ( p < 0.05). An “over-fitting” phenomenon tends to occur in BP neural networks, while RF method is more robust, providing a detection effect that is better than the BP neural networks. Thus, the application of the random forest algorithm for pest damage and multilevel dispersed variables is thus feasible and suggests that attention to the proportionality of sample data from various categories is needed when collecting data.

Forest Pest Prediction Method with Apriori Algorithm

Application of Rule-based Outlier Data Mining to Pest and Disease Forecasting

The Application of Data Mining Technique on Railway Public Security

Application of Apriori Algorithm on Data Analyses of Forest Inventory

Association Rule Mining Algorithm Based on Spark for Pesticide Transaction Data Analyses

Mining of spatial association rules in Chinese energy-saving vegetation database based on apriori algorithm

The Research of Intrusion Detection System Based on Improved Apriori Algorithm of Data Mining Association Rules

Research and application of fuzzy association rules algorithm

Research on Prediction System of Pest Distribution Based on GIS

Mining Decision Rules Based on the Improved Apriori Algorithm

Research on Data Mining Technology for Fault Correlation Analysis of Power Communication Network

Intrusion Detection Model Based on Fuzzy Theory and Association Rules

BP neural networks and random forest models to detect damage by Dendrolimus punctatus Walker

Study on Ontology-Based Apriori Algorithm Applying to Emergency Decision System

An early warning model for vegetable pests based on multidimensional data

Pest Prediction Research Based on Combining Factor Analysis and Neural Network

An implementation of improved apriori algorithm

Predicting Forest Fire Risk Based on Mining Rules with Ant-Miner Algorithm in Cloud-Rich Areas

The Diagnosis of Crop Pests and Diseases Based on the Heuristic Search

RANDOM FOREST AND OBJECTED-BASED CLASSIFICATION FOR FOREST PEST EXTRACTION FROM UAV AERIAL IMAGERY

Web Log Data Mining Based on Association Rule