Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract.

The critical components for applying the correct amount of agrochemicals are the fruit tree characteristics such as canopy height, canopy volume, and canopy cover. An unmanned aerial vehicle (UAV)-based tree canopy characteristics measurement system was developed in this study using image processing approaches. A high-resolution red-green-blue (RGB) camera was equipped with the UAV to capture images. The captured images were used to generate a digital surface model (DSM) and a digital terrain model (DTM). A tree canopy height map was generated from the subtraction of DSM and DTM. A total of 24 apple trees were targeted to measure the canopy characteristics. Region of interest (ROI) was generated across the boundary of each targeted tree. The height of all pixels within each ROI was computed separately. The pixel with maximum height was considered as the height of the respective tree. For computing canopy volume, the sum of all pixel heights from individual ROI was multiplied with the square of ground sample distance (GSD) of 5.69 mm·pixel^-1. A segmentation method was employed to calculate the canopy cover of the individual trees. The segmented canopy pixel area was divided by the total pixel area within the ROI. The results showed an average relative error of 0.2 m (6.64%) while comparing automatically measured tree height with ground measurements. For tree canopy volume, a mean absolute error of 0.25 m and a root mean square error of 0.33 m were achieved. Results of tree canopy cover measurement showed promising while compared with visual assessment. The overall results suggest that the UAV-based tree canopy characteristics measurements could be used to quantify the tree canopy characteristics to calculate the pesticide requirement for precision spraying applications in tree fruit orchards.