Biomass estimation from canopy measurements for leafy vegetables based on ultrasonic and laser sensors

This paper reports on the evaluation of a prototype sensor system embedded in a low-cost portable instrument for leaf vegetable production. The system involves circular scanning of crop canopies to identify crop biomass yield using distance sensors. The results of these scans are height profiles along an angular position from 0 to 360 degrees, which are the input for the biomass estimation. Two experiments were developed to test the performance of the system's prototype. The first experiment was conducted in a greenhouse with lettuce and kale. Biomass was estimated from the sensor system's measurements resulting in the coefficient of determination (R-2) for regression between measured and predicted biomass between 0.74 and 0.93, root mean squared error (RMSE) between 0.295 ln(g) and 0.441 ln(g), and percentage error between 25% and 55%. These values include both dry and fresh biomass for lettuce and kale. The second experiment was conducted in a spinach field of a commercial farm in Sherrington, Quebec, Canada. The R-2 values were 0.78 and 0.94. The RMSE was, in turn, 4.18 t/ha and 2.16 t/ha. While only fresh biomass was considered for the spinach case, two approaches for processing laser-based height profiles are discussed: regression of profile-representative features and inference of a canopy density function. The results indicate that the developed sensor system would be a suitable tool for rapid assessment of fresh biomass in the field. Its application would be beneficial in the process of optimizing crop management logistics, comparing the performance of different varieties of crops, and detecting potential stresses in a field environment.