DFN-PSAN: Multi-level deep information feature fusion extraction network for interpretable plant disease classification

•A novel DFN-PSAN model is proposed to identify crop diseases in natural farmland environments accurately.•The multi-level Deep Information Feature Fusion Network (DFN), which can effectively extract and fuse relevant features from different network layers, improves the localization of infected plant disease areas.•Pyramid Squeeze Attention (PSA) fuses contextual information at different scales and produces better pixel-level attention.•The construction of the PSA attention classification network (PSAN) can effectively utilize the important feature information of DFN to achieve competitive performance on complex disease symptom datasets in the field.•The t-SNE and SHAP methods enhance the transparency of the model in terms of feature clustering and discrimination of multi-category disease attention, respectively. Accurate identification of crop diseases is an effective way to promote the development of intelligent and modernized agricultural production, as well as to reduce the use of pesticides and improve crop yield and quality. Deep learning methods have achieved better performance in classifying input plant disease images. However, many plant disease datasets are often constructed from controlled scenarios, and these deep learning models may not perform well when tested in real-world agricultural environments, highlighting the challenges of transitioning to natural farm environments under the new demand paradigm of Agri 4.0. Based on the above reasons, this work proposes using a multi-level deep information feature fusion extraction network (DFN-PSAN) to achieve plant disease classification in natural field environments. DFN-PSAN adopts the YOLOv5 Backbone and Neck network as the base structure DFN and uses pyramidal squeezed attention (PSA) combined with multiple convolutional layers to design a novel classification network PSAN, which fuses and processes the multi-level depth information features output from DFN and highlights the critical regions of plant disease images with the help of pixel-level attention provided by PSA, thus realizing effective classification of multiple fine-grained plant diseases. The proposed DFN-PSAN was trained and tested on three plant disease datasets. The average accuracy and F1-score exceeded 95.27%. The PSA attention mechanism saved 26% of model parameters, achieving a competitive performance among existing related methods. In addition, this work effectively enhances the transparency of the features of the model atten