Preharvest Monitoring of Biomass Production - Engineering and Science of Biomass Feedstock Production and Provision - page 77

Fig. 4.15 UAV fl yover waypoint planning with geo-reference

University of Illinois at Urbana-Champaign [ 35 ]. The system mainly consisted of a
remote control (RC) helicopter, a multispectral camera, an IMU, a WAAS (Wide
Area Augmentation System) differential corrected GPS sensor, a single board com-
puter (SBC), a fl ight controller, a PWM (Pulse Width Modulation) switch, a wire-
less router, and a video transmitter. The camera, ADC, used a CMOS (Complementary
Metal Oxide Semiconductor) sensor with 3.2 million (2,048 × 1,036) pixels to sense
green band (520-620 nm), red band (620-750 nm), and near-infrared band (750-
950 nm) images. A lens with 8-mm focal length and maximum aperture F1.6 was
used on the camera. The camera can be triggered by the PWM switch at desired
locations. The sensors used in the IMU were a three-axis rate gyro, a three-axis
accelerometer, and a three-axis magnetometer. The SBC was used to fuse all sensors
data to estimate the UAV navigation data (altitude and position) at 50 Hz [ 36 - 38 ].

4.9.2 Aerial Image Acquisition and Analysis

The UAV-based remote sensing system is able to fl y over at certain intervals and get
the right images of the crop over the growing season with fl ying path planning func-
tion, which is indicated in Fig. 4.15 . The fl y waypoint can be calculated based on
the spatial resolution requirement and the camera parameter such as the view of
angle, focal length, and resolution.
Based on the database from tower-based remote sensing system, the growth pat-
tern of each bioenergy crop could be recognized from the daily remote sensing data
as shown in Fig. 4.16 . Thus, large-scale biomass yield prediction can be achieved
with UAV remote sensing image or biweekly satellite images.