A Method for Classifying Wood-Boring Insects for Pest Control Based on Deep Learning Using Boring Vibration Signals with Environment Noise

Forests, Journal Year: 2024, Volume and Issue: 15(11), P. 1875 - 1875

Published: Oct. 25, 2024

Wood-boring pests are difficult to monitor due their concealed lifestyle. To effectively control these wood-boring pests, it is first necessary efficiently and accurately detect presence identify species, which requires addressing the limitations of traditional monitoring methods. This paper proposes a deep learning-based model called BorerNet, incorporates an attention mechanism using limited vibration signals generated by feeding larvae. Acoustic sensors can be used collect boring from larvae emerald ash borer (EAB), Agrilus planipennis Fairmaire, 1888 (Coleoptera: Buprestidae), small carpenter moth (SCM), Streltzoviella insularis Staudinger, 1892 (Lepidoptera: Cossidae). After preprocessing steps such as clipping segmentation, Mel-frequency cepstral coefficients (MFCCs) extracted inputs for BorerNet model, with noisy real environments test set. learns input features outputs identification results. The research findings demonstrate that achieves accuracy 96.67% exhibits strong robustness generalization capabilities. Compared methods, this approach offers significant advantages in terms automation, recognition efficiency, cost-effectiveness. It enables early detection treatment pest infestations allows development targeted strategies different pests. introduces innovative technology into field tree health monitoring, enhancing ability making substantial contribution forestry-related practical applications.

Language: Английский

The Ecology, Economics, and Management of Agrilus Beetles

Current Forestry Reports, Journal Year: 2024, Volume and Issue: 10(6), P. 487 - 509

Published: Sept. 24, 2024

Language: Английский