Cotton boll distribution is a critical phenotypic trait that represents the plant's response to its environment. Accurate quantification of boll distribution provides valuable information for breeding cultivars with high yield and fiber quality. Manual methods for boll mapping are time-consuming and labor-intensive. We evaluated the application of Lidar point cloud and RGB image data in boll detection and distribution and yield estimation. Lidar data was acquired at 15 m using a DJI Matrice 300 RTK and RGB images were acquired using a DJI Phantom 4 RTK. We assessed different classification methods, including a region-based algorithm and a density-based clustering method to detect cotton bolls and generate the boll distribution and estimate cotton yield.