Abstract: To address the limitations of traditional phenotyping in crops, including low efficiency, limited accuracy and destructive sampling, this paper aimed to review the applications of 3D point cloud processing technology in crop high-throughput phenotyping, and systematically survey the extended application status of phenotype-driven smart agriculture. First, the paper summarized crop 3D point cloud acquisition technologies including LiDAR, 3D reconstruction technology based on multi-camera stereo vision, and depth cameras, and analyzed their robustness in complex agricultural scenarios. Second, the evolution of crop 3D point cloud segmentation algorithms which shifted from “conventional handcrafted feature engineering + machine-learning-based regression” paradigm towards “deep learning” was systematically reviewed, with emphasis placed on the advantages of deep learning models (such as PointNet++ and Transformer) in addressing non-rigid deformation, organ similarity and complex structure segmentation. Third, this paper elaborated on the current status of crop high-throughput phenotyping, covering digital twin-based phenotypic trait temporal tracking technology and mainstream indoor and field highly integrated phenotyping platforms worldwide. Finally, smart extended applications of phenotypic information were further reviewed, covering crop growth monitoring, early stress diagnosis, and the operational status of phenotypic-information-driven intelligent agricultural machinery in autonomous navigation, precision spraying, intelligent pruning and automated harvesting. By systematically organizing existing technologies and methods, this paper to reveals their advantages and limitations, provides references for innovative applications in crop phenotyping and intelligent agricultural equipment, and proposes potential future research directions.