In the modern field of biological pest control, especially in the realm of insect population monitoring, deep learning methods have made further advancements. However, due to the small size and elusive nature of insects, visual detection is often impractical. In this context, the recognition of insect sound features becomes crucial. In our study, we introduce a classification module called the “dual-frequency and spectral fusion module (DFSM)”, which enhances the performance of transfer learning models in audio classification tasks. Our approach combines the efficiency of EfficientNet with the hierarchical design of the Dual Towers, drawing inspiration from the way the insect neural system processes sound signals. This enables our model to effectively capture spectral features in insect sounds and form multiscale perceptions through inter-tower skip connections. Through detailed qualitative and quantitative evaluations, as well as comparisons with leading traditional insect sound recognition methods, we demonstrate the advantages of our approach in the field of insect sound classification. Our method achieves an accuracy of 80.26% on InsectSet32, surpassing existing state-of-the-art models by 3 percentage points. Additionally, we conducted generalization experiments using three classic audio datasets. The results indicate that DFSM exhibits strong robustness and wide applicability, with minimal performance variations even when handling different input features.

The application of 3D digital models to high-throughput plant phenotypic analysis is a research hotspot nowadays. Traditional methods, such as manual measurement and laser scanning, have high costs, and multi-view, unsupervised reconstruction methods are still blank in the field of crop research. It is challenging to obtain a high-quality 3D crop surface feature composition for 3D reconstruction. In this paper, we propose a wheat point cloud generation and 3D reconstruction method based on SfM and MVS using sequential wheat crop images. Firstly, the camera intrinsics and camera extrinsics of wheat were estimated using a structure-from-motion system with feature maps, which effectively solved the problem of camera point location design. Secondly, we proposed the ReC-MVSNet, which integrates the heavy parametric structure into the point cloud 3D reconstruction network, overcoming the difficulty of capturing complex features via the traditional MVS model. Through experiments, it was shown that this research method achieves non-invasive reconstruction of the 3D phenotypic structure of realistic objects, the accuracy of the proposed model was improved by nearly 43.3%, and the overall value was improved by nearly 14.3%, which provided a new idea for the development of virtual 3D digitization.