Rice leaf disease detection based on bidirectional feature attention pyramid network with YOLO v5 model

• Published in: Environmental Research Communications Vol. 5; no. 6; pp. 65014 - 65031
• Main Authors: Kumar, V Senthil, Jaganathan, M, Viswanathan, A, Umamaheswari, M, Vignesh, J
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.06.2023
• Subjects: agricultural industry; Agrochemicals; Artificial neural networks; bidirectional feature attention pyramid network; Cereal crops; Computer networks; Crop diseases; Crop production; Datasets; DenseNet-201; Disease detection; economic development; Environmental degradation; Feature maps; Image segmentation; instance segmentation; Leaves; Medical imaging; Neural networks; Object recognition; Pesticides; Plant diseases; Rice; rice leaf disease detection; Vectors; YOLO v5 network
• ISSN: 2515-7620; 2515-7620
• DOI: 10.1088/2515-7620/acdece

To ensure higher quality, capacity, and production of rice, it is vital to diagnose rice leaf disease in its early stage in order to decrease the usage of pesticides in agriculture which in turn avoids environmental damage. Hence, this article presents a Multi-scale YOLO v5 detection network to detect and classify the rice crop disease in its early stage. The experiment is initially started by pre-processing the rice leaf images obtained from the RLD dataset, after which data set labels are created, which are then divided into train and test sets. DenseNet-201 is used as the backbone network and depth-aware instance segmentation is used to segment the different regions of rice leaf. Moreover, the proposed Bidirectional Feature Attention Pyramid Network (Bi-FAPN) is used for extracting the features from the segmented image and also enhances the detection of diseases with different scales. Furthermore, the feature maps are identified in the detection head, where the anchor boxes are then applied to the output feature maps to produce the final output vectors by the YOLO v5 network. The subset of channels or filters is pruned from different layers of deep neural network models through the principled pruning approach without affecting the full framework performance. The experiments are conducted with RLD dataset with different existing networks to verify the generalization ability of the proposed model. The effectiveness of the network is evaluated based on various parameters in terms of average precision, accuracy, average recall, IoU, inference time, and F1 score, which are achieved at 82.8, 94.87, 75.81, 0.71, 0.017, and 92.45 respectively.