Abstract: 【Objective】 This study aims to make use of the Unmanned Aerial Vehicle (UAV) remote sensing image technology platform to obtain the phenotypic information of field crop height through rapid, non-destructive and high-throughput methods, which is of great significance for the growth monitoring and yield prediction of cotton varieties (lines). 【Method】 In this paper, UAV is equipped with high-definition digital camera to form a low-altitude remote sensing platform and low-altitude UAV remote sensing technology was used to obtain regional high-precision remote sensing images for 110 cotton varieties (lines) in the flowering and boll period planted in the cotton breeding test field base of Xinjiang Agricultural University, Shawan County, Xinjiang, and measured the actual plant height of the ground manually. Firstly, we used the Pix4D mapper and the high-definition digital image to generate the Digital Surface Model (DSM) and the Digital Orthophoto Map (DOM) of the study area. Then, based on DOM and high-definition DSM, the Kriging interpolation method was used to generate the discrete Digital Elevation Model (DEM) of the study area. The Crop Height Model (CHM) of the study area was obtained by the difference between the high-definition DSM and the discrete DEM of the study area. Finally, regression analysis was carried out using the measured plant height (H) of different varieties (lines) and the CHM of the extracted cotton. 【Results】 The DOM could quickly and non-destructively monitor the growth, leaf color trait differences and distribution of cotton varieties (lines) at the flower-boll stage. According to the DEM and cotton plant height distribution maps extracted by DSM and Kriging interpolation, the overall terrain of the study area was relatively flat, and the height difference was only 0.5 m. The constructed plant height model R² reached 0.8469, and the verification model R² also reached 0.7581, indicating that the drone equipped with a digital camera had a good applicability for cotton plant height measurement. 【Conclusion】 The DEM generated by DOM, DSM and Kriging interpolation method generated by UAV images is used to extract cotton plant height (CHM) at flowering stage, which reaches relatively high accuracy. It provides a new research method for large-scale observation of cotton field plant height, which proves that the experimental research method is feasible.