Monitoring fruit tree flowering information in the open world is more crucial than in the research-oriented environment for managing agricultural production to increase yield and quality. This work presents a transformer-based flowering period monitoring approach in an open world in order to better monitor the whole blooming time of modern standardized orchards utilizing IoT technologies. This study takes images of flowering apple trees captured at a distance in the open world as the research object, extends the dataset by introducing the Slicing Aided Hyper Inference (SAHI) algorithm, and establishes an S-YOLO apple flower detection model by substituting the YOLOX backbone network with Swin Transformer-tiny. The experimental results show that S-YOLO outperformed YOLOX-s in the detection accuracy of the four blooming states by 7.94%, 8.05%, 3.49%, and 6.96%. It also outperformed YOLOX-s by 10.00%, 9.10%, 13.10%, and 7.20% for mAP_ALL, mAP_S, mAP_M, and mAP_L, respectively. By increasing the width and depth of the network model, the accuracy of the larger S-YOLO was 88.18%, 88.95%, 89.50%, and 91.95% for each flowering state and 39.00%, 32.10%, 50.60%, and 64.30% for each type of mAP, respectively. The results show that the transformer-based method of monitoring the apple flower growth stage utilized S-YOLO to achieve the apple flower count, percentage analysis, peak flowering time determination, and flowering intensity quantification. The method can be applied to remotely monitor flowering information and estimate flowering intensity in modern standard orchards based on IoT technology, which is important for developing fruit digital production management technology and equipment and guiding orchard production management.

Automatic plant phenotype measurement technology based on the rapid and accurate reconstruction of maize structures at the seedling stage is essential for the early variety selection, cultivation, and scientific management of maize. Manual measurement is time-consuming, laborious, and error-prone. The lack of mobility of large equipment in the field make the high-throughput detection of maize plant phenotypes challenging. Therefore, a global 3D reconstruction algorithm was proposed for the high-throughput detection of maize phenotypic traits. First, a self-propelled mobile platform was used to automatically collect three-dimensional point clouds of maize seedling populations from multiple measurement points and perspectives. Second, the Harris corner detection algorithm and singular value decomposition (SVD) were used for the pre-calibration single measurement point multi-view alignment matrix. Finally, the multi-view registration algorithm and iterative nearest point algorithm (ICP) were used for the global 3D reconstruction of the maize seedling population. The results showed that the R² of the plant height and maximum width measured by the global 3D reconstruction of the seedling maize population were 0.98 and 0.99 with RMSE of 1.39 cm and 1.45 cm and mean absolute percentage errors (MAPEs) of 1.92% and 2.29%, respectively. For the standard sphere, the percentage of the Hausdorff distance set of reconstruction point clouds less than 0.5 cm was 55.26%, and the percentage was 76.88% for those less than 0.8 cm. The method proposed in this study provides a reference for the global reconstruction and phenotypic measurement of crop populations at the seedling stage, which aids in the early management of maize with precision and intelligence.