RESUMO
Maize (Zea mays L.) is one of the most important crops, influencing food production and even the whole industry. In recent years, global crop production has been facing great challenges from diseases. However, most of the traditional methods make it difficult to efficiently identify disease-related phenotypes in germplasm resources, especially in actual field environments. To overcome this limitation, our study aims to evaluate the potential of the multi-sensor synchronized RGB-D camera with depth information for maize leaf disease classification. We distinguished maize leaves from the background based on the RGB-D depth information to eliminate interference from complex field environments. Four deep learning models (i.e., Resnet50, MobilenetV2, Vgg16, and Efficientnet-B3) were used to classify three main types of maize diseases, i.e., the curvularia leaf spot [Curvularia lunata (Wakker) Boedijn], the small spot [Bipolaris maydis (Nishik.) Shoemaker], and the mixed spot diseases. We finally compared the pre-segmentation and post-segmentation results to test the robustness of the above models. Our main findings are: 1) The maize disease classification models based on the pre-segmentation image data performed slightly better than the ones based on the post-segmentation image data. 2) The pre-segmentation models overestimated the accuracy of disease classification due to the complexity of the background, but post-segmentation models focusing on leaf disease features provided more practical results with shorter prediction times. 3) Among the post-segmentation models, the Resnet50 and MobilenetV2 models showed similar accuracy and were better than the Vgg16 and Efficientnet-B3 models, and the MobilenetV2 model performed better than the other three models in terms of the size and the single image prediction time. Overall, this study provides a novel method for maize leaf disease classification using the post-segmentation image data from a multi-sensor synchronized RGB-D camera and offers the possibility of developing relevant portable devices.
RESUMO
Accurate and rapid identification of the effective number of panicles per unit area is crucial for the assessment of rice yield. As part of agricultural development, manual observation of effective panicles in the paddy field is being replaced by unmanned aerial vehicle (UAV) imaging combined with target detection modeling. However, UAV images of panicles of curved hybrid Indica rice in complex field environments are characterized by overlapping, blocking, and dense distribution, imposing challenges on rice panicle detection models. This paper proposes a universal curved panicle detection method by combining UAV images of different types of hybrid Indica rice panicles (leaf-above-spike, spike-above-leaf, and middle type) from four ecological sites using an improved You Only Look Once version 4 (YOLOv4) model. MobileNetv2 is used as the backbone feature extraction network based on a lightweight model in addition to a focal loss and convolutional block attention module for improved detection of curved rice panicles of different varieties. Moreover, soft non-maximum suppression is used to address rice panicle occlusion in the dataset. This model yields a single image detection rate of 44.46 FPS, and mean average precision, recall, and F1 values of 90.32%, 82.36%, and 0.89%, respectively. This represents an increase of 6.2%, 0.12%, and 16.24% from those of the original YOLOv4 model, respectively. The model exhibits superior performance in identifying different strain types in mixed and independent datasets, indicating its feasibility as a general model for detection of different types of rice panicles in the heading stage.
