Evaluating RGB Imaging and Multispectral Active and Hyperspectral Passive Sensing for Assessing Early Plant Vigor in Winter Wheat

Abstract

Plant vigor is an important trait of field crops at early growth stages, influencing weed suppression, nutrient and water use efficiency and plant growth. High-throughput techniques for its evaluation are required and are promising for nutrient management in early growth stages and for detecting promising breeding material in plant phenotyping. However, spectral sensing for assessing early plant vigor in crops is limited by the strong soil background reflection. Digital imaging may provide a low-cost, easy-to-use alternative. Therefore, image segmentation for retrieving canopy cover was applied in a trial with three cultivars of winter wheat (Triticum aestivum L.) grown under two nitrogen regimes and in three sowing densities during four early plant growth stages (Zadok’s stages 14–32) in 2017. Imaging-based canopy cover was tested in correlation analysis for estimating dry weight, nitrogen uptake and nitrogen content. An active Greenseeker sensor and various established and newly developed vegetation indices and spectral unmixing from a passive hyperspectral spectrometer were used as alternative approaches and additionally tested for retrieving canopy cover. Before tillering (until Zadok’s stage 20), correlation coefficients for dry weight and nitrogen uptake with canopy cover strongly exceeded all other methods and remained on higher levels (R² > 0.60***) than from the Greenseeker measurements until tillering. From early tillering on, red edge based indices such as the NDRE and a newly extracted normalized difference index (736 nm; ~794 nm) were identified as best spectral methods for both traits whereas the Greenseeker and spectral unmixing correlated best with canopy cover. RGB-segmentation could be used as simple low-cost approach for very early growth stages until early tillering whereas the application of multispectral sensors should consider red edge bands for subsequent stages.

Deep Phenotyping of Yield-Related Traits in Wheat

“…This seasonal pattern is in line with the previous GY estimation across the full trial data [51] and with the spectral estimation of the N traits [47]. Concerning the seasonal pattern in Cont_SD1, the SVI*date interaction ( Figure 5) confirms the relative weakness of water band indices during the early phase [49] for 'planar', un-saturated canopies, being in line with [47]. However, already at tillering, the differentiation with a good performance of the water band indices and of most RE indices, but the failure of NIR/VIS indices, possibly indicates the advantage of a stronger sensitivity of the former groups for overcoming saturation in dense canopies, as also observed for wheat breeding lines [26], N fertilization in wheat cultivars [51], drought-stressed wheat [39] and barley [52].…”

“…Therefore, this study included multiple DM traits that contribute to grain filling for elucidating the GY~SVI relationships, and multiple measurement dates, given that the trial treatments influenced the traits in different phases [25]. Because of the effects of sowing date on the early development in terms of DM traits, canopy cover, and SVIs [49], and the positive yield effect of early sowing (SD1) [25], moderate relationships were found already during leaf development and tillering, but not in the datasets without varying sowing dates, corroborating results in a breeding population [26]. Moreover, given that R 2 -values again decreased to a minimum until ear emergence/anthesis, as also found in [26], this early discrimination confirms the spectral detection of early vigor [49,50], but should not be generalized for the prediction of grain yield.…”

“…Because of the effects of sowing date on the early development in terms of DM traits, canopy cover, and SVIs [49], and the positive yield effect of early sowing (SD1) [25], moderate relationships were found already during leaf development and tillering, but not in the datasets without varying sowing dates, corroborating results in a breeding population [26]. Moreover, given that R 2 -values again decreased to a minimum until ear emergence/anthesis, as also found in [26], this early discrimination confirms the spectral detection of early vigor [49,50], but should not be generalized for the prediction of grain yield. This seasonal pattern is in line with the previous GY estimation across the full trial data [51] and with the spectral estimation of the N traits [47].…”

Monitoring Plant Status and Fertilization Strategy through Multispectral Images

“…These sensors are based on multispectral images and use vegetation indices to obtain the best correlation with the nutrient status of the arable crops [12][13][14].…”

Temporal and Organ-specific Responses in NUE Traits to N Fertilization, Fungicide Intensity and Early Sowing in Winter Wheat Cultivars

“…The winter wheat cultivars comprised three high-performance commercial German line cultivars and three hybrid cultivars, registered and released by the national authorities. Additionally, three other cultivars were included in all main plots but neglected due to the previous sampling during early plant development [48], and another 11 cultivars, included only in Cont and not destructively sampled. The cultivars include three quality groups for winter wheat (E: Highest baking, A: High baking quality and B: Bread wheat) and differ in yield components, disease susceptibility and phenology ( Table 1).…”