Abstract: Abstract: Phenotypic information of individual corn plant is of great significance for the study on the rules of high yield and high efficiency growth of corn and the genotype determination on corn genetic breeding. However, the traditional method used to extract phenotypic information is time-consuming, inefficient and subjective, which cannot meet the needs of current corn research. For the past few years, the development of remote sensing technology has laid a foundation for the rapid and efficient extraction of crop phenotypic information. Terrestrial laser scanning (TLS) has the characteristics of high precision, non-destructive measuring and multi-parameter-extraction, which is suitable for high-precision phenotypic analysis of crop breeding. The structure of the whole plant can be measured on a millimeter scale, and the data analysis process enables the multiple morphological plant parameters can be derived simultaneously in a single laser scan. Therefore, a TLS-based method for extracting phenotypic information of corn grown in field was proposed in this paper. Field experiments were carried out in Baoding City, Hebei province in 2016. In order to ensure the integrity of data, three stations were set up to scan the corn in the target sample, and Trimble TX8 3D scanner was used to obtain high-precision 3D point cloud data of corn in tasseling stage. Generating the required phenotypic parameters from the massive point cloud data was performed in a multi-step process, including multi-station data registration, removal of ground points and other invalid points, denoising, separation of individual corn from the corn population, resampling and separation of stem and leaves. Data registration was done using the Trimble Realworks software. Points represented invalid points such as ground points were removed using the adaptive triangulated irregular network (TIN) algorithm provided by version 016.004 of the Terrasolid software. This algorithm creates a sparse TIN from seed points and then iteratively densifies the TIN until all noise points are identified and removed. The denoising was done by Geomagic studio 2014 software. Individual corn plant was separated from the corn population using cloudcompare. Then resampling was performed. Finally, corn leaf points were separated from stalk points using the difference of normals (DoN) method. Providing a description of corn plants was helpful to simplify data processing without affecting the underlying point cloud and to achieve objective parameterization of growth state. Therefore, leaf model was constructed using triangle meshes, and corn leaf skeleton was extracted using Geomagic Studio 2014. Then the actual leaf area was calculated by calculating the sum of the areas of all triangular meshes. Leaf length was extracted by calculating the euclidean distance between the leaf points, and the leaf inclination angle was obtained by piecewise fitting with least square method based on leaf skeleton model. The plant height was calculated from the corn plant point cloud. To compare the values obtained by the proposed method in this paper and those obtained by manual measurement, the regression analysis was done with the root mean square error and calculated mean absolute percentage errors. Results showed that the determination coefficients (R2) of actual leaf area, leaf length, plant height and leaf inclination angle were 0.99, 0.99, 0.96 and 0.94, respectively, the root mean square error were 12.69 cm2, 1.31 cm, 1.30 cm and 5.12° respectively, and the average relative errors were 2.38%, 1.32%, 0.61% and 8.96% respectively. Therefore, the method proposed in this paper can be used to extract the phenotypic information for individual corn. Advances in high-throughput phenotypes will be conceivable through the combination of digital imaging techniques and further automated data analysis steps. This will help speed up the process of plant breeding.