Traditional deep learning methods employing 2D images can only classify healthy and unhealthy seedlings; consequently, this study proposes a method by which to further classify healthy seedlings into primary seedlings and secondary seedlings and finally to differentiate three classes of seedling through a 3D point cloud for the detection of useful eggplant seedling transplants. Initially, RGB images of three types of substrate-cultivated eggplant seedlings (primary, secondary, and unhealthy) were collected, and healthy and unhealthy seedlings were classified using ResNet50, VGG16, and MobilNetV2. Subsequently, a 3D point cloud was generated for the three seedling types, and a series of filtering processes (fast Euclidean clustering, point cloud filtering, and voxel filtering) were employed to remove noise. Parameters (number of leaves, plant height, and stem diameter) extracted from the point cloud were found to be highly correlated with the manually measured values. The box plot shows that the primary and secondary seedlings were clearly differentiated for the extracted parameters. The point clouds of the three seedling types were ultimately classified directly using the 3D classification models PointNet++, dynamic graph convolutional neural network (DGCNN), and PointConv, in addition to the point cloud complementary operation for plants with missing leaves. The PointConv model demonstrated the best performance, with an average accuracy, precision, and recall of 95.83, 95.83, and 95.88%, respectively, and a model loss of 0.01. This method employs spatial feature information to analyse different seedling categories more effectively than two-dimensional (2D) image classification and three-dimensional (3D) feature extraction methods. However, there is a paucity of studies applying 3D classification methods to predict useful eggplant seedling transplants. Consequently, this method has the potential to identify different eggplant seedling types with high accuracy. Furthermore, it enables the quality inspection of seedlings during agricultural production.

For different broccoli materials, it used to be necessary to manually plant in a large area for the investigation of flower ball information, and this method is susceptible to subjective influence, which is not only time-consuming and laborious but may also cause some damage to the broccoli in the process of investigation. Therefore, the rapid and nondestructive monitoring of flower heads is key to acquiring high-throughput phenotypic information on broccoli crops. In this study, we used an unmanned aerial vehicle (UAV) to acquire hundreds of images of field-grown broccoli to evaluate their flower head development rate and sizes during growth. First, YOLOv5 and YOLOv8 were used to complete the position detection and counting statistics at the seedling and heading stages. Then, UNet, PSPNet, DeepLabv3+, and SC-DeepLabv3+ were used to segment the flower heads in the images. The improved SC-DeepLabv3+ model excelled in segmenting flower heads, showing Precision, reconciled mean F1-score, mean intersection over union, and mean pixel accuracy values of 93.66%, 95.24%, 91.47%, and 97.24%, respectively, which were 0.57, 1.12, 1.16, and 1.70 percentage points higher than the respective values achieved with the DeepLabv3+ model. Flower head sizes were predicted on the basis of the pixel value of individual flower heads and ground sampling distance, yielding predictions with an R² value of 0.67 and root-mean-squared error of 1.81 cm. Therefore, the development rate and sizes of broccoli flower heads during growth were successively estimated and calculated. Compared with the existing technology, it greatly improves work efficiency and can help to obtain timely information on crop growth in the field. Our methodology provides a convenient, fast, and reliable way for investigating field traits in broccoli breeding.