On-site monitoring of crop growth throughout the growing season plays an important role in assessing overall crop conditions, determining when to irrigate, and forecasting potential yields. Conventional crop monitoring methods, involving manual sampli...
On-site monitoring of crop growth throughout the growing season plays an important role in assessing overall crop conditions, determining when to irrigate, and forecasting potential yields. Conventional crop monitoring methods, involving manual sampling and laboratory analysis, are time consuming and labor-intensive, thereby requiring fast measurements with a high sampling intensity. In our previous studies, both mathematical modeling based on ExG-based vegetation fraction (VF) and SfM-estimated plant height (PH), and validation testing of the developed models were conducted as a means of measuring various biophysical properties of Chinese cabbage and white radish using UAV-RGB imagery. This study reports on automatic temporal and spatial yield mapping and kriging of the two different crops grown in a farmer’s field using crop automatic identification algorithms. Remotely sensed RGB images were collected on an approximately 1-week interval from September 2017 to November 2017 using an UAV platform flying at 2 m/s at 20 m AGL. In this study, an image preprocessing algorithm that could automatically perform yield mapping of the two crops was developed by applying a technique of identifying the center of each crop in RGB images. As a result, fresh weight maps and kriging of Chinese cabbage and white radish were automatically generated that effectively showed temporal and spatial variability in their growth status. Spatial yield mapping can be automatically performed on the desired date for all growth parameters. This indicates that yield maps generated using the UAV RGB images in conjunction with the developed prediction models could be used to evaluate the potential yields of the two crops prior to harvest.