The Smart Plant Protection Innovation Team of the Institute of Plant Protection, Chinese Academy of Agricultural Sciences (IPPCAAS), recently published a research paper titled "Precise Extraction of Targeted Apple Tree Canopy with YOLO-Fi Model for Advanced UAV Spraying Plans" in the prestigious journal Computers and Electronics in Agriculture . The study introduces the YOLO-Fi algorithm model, designed for multi-objective tasks such as orchard tree recognition, localization, and segmentation, enabling an integrated technical solution for variable spraying by plant protection UAVs.

Accurate analysis of individual orchard tree canopy information and precise navigation and spraying operations of plant protection machinery are critical for smart orchard management. However, simultaneously detecting, localizing, and segmenting tree canopies in complex orchard environments remains highly challenging. This study proposes an integrated framework based on UAV data and deep learning algorithms to precisely acquire apple tree canopy information for UAV-targeted variable spraying. First, the mRMR (Max-Relevance and Min-Redundancy) algorithm was used to select three features (RVI, NDVI, SAVI) to enhance images, enabling tree canopies to stand out from the background. Secondly, leveraging the labeled dataset, the YOLO-Fi model was developed. Using this optimal model, precise detection, localization, and segmentation of fruit trees in the experimental area were conducted. The model demonstrated superior performance, achieving optimal results (FPS = 370, mAP50-95(B) = 0.862, mAP50-95(M) = 0.723, and MIoU = 0.749). Subsequently, based on the segmented areas of the fruit tree canopies, a variable spraying prescription map was generated, contributing to a 47.92% reduction in spraying volume compared to direct spraying. Finally, the ant colony algorithm was employed to design the shortest path for the plant protection UAV to traverse over each fruit tree within the experimental area, leading to a 2.04% reduction in distance compared to the conventional UAV flight path.

This study provides an integrated solution for orchard tree canopy monitoring, analysis, localization, navigation, and precise pesticide application, offering strong technical support for UAV-based smart orchard management.

IPPCAAS is the primary institution for this research. Ph.D. candidate Wei Peng is the first author, with Professors Yuan Huizhu and Yan Xiaojing serving as co-corresponding authors. The study was supported by the National Key R&D Program (2022YFD2001402).