Phongchanmixay (2020) Field phenotyping of plant height in an upland rice field in Laos using low-cost small unmanned aerial vehicles (UAVs),

In this study, we propose a method for discriminating crops/weeds in upland rice fields using a commercial unmanned aerial vehicles (UAVs) and red-green-blue (RGB) cameras with the simple linear iterative clustering (SLIC) algorithm and random forest (RF) classifier. In the SLIC-RF algorithm, we evaluated different combinations of input features: three color spaces (RGB, hue-saturation-brightness [HSV], CIE-L*a*b), canopy height model (CHM), spatial texture (Texture) and four vegetation indices (VIs) (excess green [ExG], excess red [ExR], green-red vegetation index [GRVI] and color index of vegetation extraction [CIVE]). Among the color spaces, the HSV-based SLIC-RF model showed the best performance with the highest out-of-bag (OOB) accuracy (0.904). The classification accuracy was improved by the combination of HSV with CHM, Texture, ExG, or CIVE. The highest OOB accuracy (0.915) was obtained from the HSV+Texture combination. The greatest errors from the confusion matrix occurred in the classification between crops and weeds, while soil could be classified with a very high accuracy. These results suggest that with the SLIC-RF algorithm developed in this study, rice and weeds can be discriminated by consumer-grade UAV images with acceptable accuracy to meet the needs of site-specific weed management (SSWM) even in the early growth stages of small rice plants..

30% in the northern part (Inoue et al. 2005). Recent estimates indicated that 17% of the Laotian people still live in S&B agricultural areas, which cover 29% of the country (Messerli et al. 2009). Although the overall time trend variations in the extent of S&B agriculture have exhibited a gradual decrease at the national level, it still persists or has increased in certain rural areas, where the local poverty rate is considerably higher than the national JARQ 51 (4), 309-318 (2017) AbstractAdaptability to a wide range of environmental factors is a key for achieving stable production in the slash-and-burn (S&B) agriculture of mountainous Laos, where soil varies widely in productivity. Adaptability assessment based on the genotype by environment (GxE) interaction of grain yields in this study entailed an investigation of the yield performance of maize, Job's tears, and seven varieties of upland rice including improved variety B6144F-MR-6-0-0 (B6144), at eleven locations under rainfed upland conditions. Across the eleven locations, the mean yields of upland rice ranged widely from 56 to 583 g m -2 . While the GxE interaction was significant, one improved indica variety (B6144) produced high grain yields at all eleven locations. Under low-yielding conditions (56 to 205 g m -2 of upland rice), four indica varieties consistently performed with more stable yield performance than three tropical japonica varieties, the poor adaptation of which was attributed to reduced grain number per panicle and the grain-filling ratio. Under moderate-and high-yielding conditions (284 to 583 g m -2), two semi-dwarf varieties -Tampi (tropical japonica) and B6144 (indica) -exhibited the highest productivity due to a higher harvest index (0.40-0.41) compared with the others (0.28 -0.34), but B6144 likely exhibited lodging signs under fertile soil conditions, where the mean yields were above 360 g m -2 . The GxE interaction effect was highly significant among the three upland crops; relative to upland rice, maize was particularly better adapted to moderate-and high-yielding conditions, whereas Job's tears adapted better to low-yielding conditions. Job's tears exhibited stable yield performance across all eleven locations. In contrast, maize had higher yield compared with all upland rice varieties under the adapted conditions, although the upland rice varieties had higher yields compared with maize under unadapted conditions. In conclusion, Job's tears or indica varieties are recommended for growth under low-yielding conditions, and maize or semi-dwarf cultivars are recommended under high-yielding conditions. And with possibly more lodging resistance, the improved indica variety B6144 could be the ideal variety adaptable to a wide range of soil fertility.