Predicting grain protein content of field-grown winter wheat with satellite images and partial least square algorithm

Tan, Changyin; Zhou, Xinxing; Zhang, Pengpeng; Wang, Zhixiang; Wang, Dunliang; Guo, Wenshan; Fei, Yun

doi:10.1371/journal.pone.0228500

Cited by 27 publications

(16 citation statements)

References 51 publications

(60 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Partial Least Square Regression (PLSR) ( 39 , 40 ) is a multivariate statistical procedure to build explanatory and predictive models to analyze multiple-response (dependent) and multiple explanatory (independent) variables, where high multicollinearity in small sample size ceases reliable conclusions due to classical regression solution. The algorithm was applied using XL stat (trial version).…”

Section: Methodsmentioning

confidence: 99%

Inferring Relationship of Blood Metabolic Changes and Average Daily Gain With Feed Conversion Efficiency in Murrah Heifers: Machine Learning Approach

et al. 2020

View full text Add to dashboard Cite

Machine learning algorithms were employed for predicting the feed conversion efficiency (FCE), using the blood parameters and average daily gain (ADG) as predictor variables in buffalo heifers. It was observed that isotonic regression outperformed other machine learning algorithms used in study. Further, we also achieved the best performance evaluation metrics model with additive regression as the meta learner and isotonic regression as the base learner on 10-fold cross-validation and leaving-one-out cross-validation tests. Further, we created three separate partial least square regression (PLSR) models using all 14 parameters of blood and ADG as independent (explanatory) variables and FCE as the dependent variable, to understand the interactions of blood parameters, ADG with FCE each by inclusion of all FCE values (i), only higher FCE values (negative RFI) (ii), and inclusion of only lower FCE (positive RFI) values (iii). The PLSR model including only the higher FCE values was concluded the best, based on performance evaluation metrics as compared to PLSR models developed by inclusion of the lower FCE values and all types of FCE values. IGF1 and its interactions with the other blood parameters were found highly influential for higher FCE measures. The strength of the estimated interaction effects of the blood parameter in relation to FCE may facilitate understanding of intricate dynamics of blood parameters for growth.

show abstract

Section: Methodsmentioning

confidence: 99%

Inferring Relationship of Blood Metabolic Changes and Average Daily Gain With Feed Conversion Efficiency in Murrah Heifers: Machine Learning Approach

et al. 2020

View full text Add to dashboard Cite

show abstract

“…S öderstr öm et al ( 2010) used a 45band spectral sensor in combination with satellite imagery to predict PC in barley, achieving reasonable PC prediction on a field level, although it was considerably less accurate when applied across multiple years and locations. Zhao et al (2019), Tan et al (2020), andXu et al (2020) predicted the PC of winter wheat using satellite data, with Zhao et al (2019) also using hyperspectral data derived from an unmanned aerial vehicle (UAV). In all of these studies, PC was predicted with good to high accuracy.…”

Section: Introductionmentioning

confidence: 99%

Preharvest phenotypic prediction of grain quality and yield of durum wheat using multispectral imaging

et al. 2022

View full text Add to dashboard Cite

Durum wheat is an important cereal that is widely grown in the Mediterranean basin. In addition to high yield, grain quality traits are of high importance for farmers. The strong influence of climatic conditions makes the improvement of grain quality traits, like protein content, vitreousness, and test weight, a challenging task. Evaluation of quality traits post-harvest is time-and labor-intensive and requires expensive equipment, such as near-infrared spectroscopes or hyperspectral imagers. Predicting not only yield but also important quality traits in the field before harvest is of high value for breeders aiming to optimize resource allocation. Implementation of efficient approaches for trait prediction, such as the use of high-resolution spectral data acquired by a multispectral camera mounted on unmanned aerial vehicles (UAVs), needs to be explored. In this study, we have acquired multispectral image data with an 11-band multispectral camera mounted on a UAV and analyzed the data with machine learning (ML) models to predict grain yield and important quality traits in breeding micro-plots. Combining 11-band multispectral data for 34 cultivars and 16 environments allowed to develop ML models with good prediction capability. Applying the trained models to test sets explained a considerable degree of phenotypic variance with good accuracy showing r squared values of 0.84, 0.69, 0.64, and 0.61 and normalized root mean squared errors of 0.17, 0.07, 0.14, and 0.03 for grain yield, protein content, vitreousness, and test weight, respectively.

show abstract

“…The spatial variability of GPC has been successfully estimated from NDVI and other vegetation indices using different remote sensing platforms [19][20][21][22][23], especially during latter stages of crop development, namely anthesis [24,25]. Wheat yield is often more strongly related to vegetation indices that are integrated across the growing-season to capture the full period of canopy development and thereby crop carbon uptake [26][27][28].…”

Section: Introductionmentioning

confidence: 99%

The spatial dynamics of wheat yield and protein content at the field scale

Stoy¹,

Khan²,

Wipf³

et al. 2021

Preprint

View full text Add to dashboard Cite

Wheat is a staple crop that is critical for feeding a hungry and growing planet, but its nutritive value has declined as global temperatures have warmed. The price offered to producers depends not only on yield but also grain protein content (GPC), which are often negatively related at the field scale but can positively covary depending in part on management strategies, emphasizing the need to predict their variability within individual fields. We measured yield and GPC in a winter wheat field in Sun River, Montana, USA and tested the ability of normalized difference vegetation index (NDVI) measurements from an unpiloted aerial vehicle (UAV) on spatial scales of ~10 cm and from Landsat on spatial scales of 30 m to predict them. Landsat observations were poorly related to wheat measurements. A multiple linear model using information from four (three) UAV flyovers was selected as the most parsimonious and predicted 26% (40%) of the variability in wheat yield (GPC). We sought to understand the optimal spatial scale for interpreting UAV observations given that the ~ 10 cm pixels yielded more than 12 million measurements at far finer resolution than the 12 m scale of the harvester. The variance in NDVI observations was ‘averaged out’ at larger pixel sizes but only ~ 20% of the total variance was averaged out at the spatial scale of the harvester on some measurement dates. Spatial averaging to the scale of the harvester also made little difference in the total information content of NDVI fit using Beta distributions as quantified using the Kullback-Leibler divergence. Radially-averaged power spectra of UAV-measured NDVI revealed relatively steep power law relationships with exponentially less variance at finer spatial scales. Results suggest that larger pixels can reasonably capture the information content of within-field NDVI, but the 30 m Landsat scale is too coarse to describe some of the key features of the field, which are consistent with topography, historic management practices, and edaphic variability. Future research should seek to determine an ‘optimum’ spatial scale for NDVI observations that minimizes effort (and therefore cost) while maintaining the ability of producers to make management decisions that positively impact yield and GPC.

show abstract

Predicting grain protein content of field-grown winter wheat with satellite images and partial least square algorithm

Cited by 27 publications

References 51 publications

Inferring Relationship of Blood Metabolic Changes and Average Daily Gain With Feed Conversion Efficiency in Murrah Heifers: Machine Learning Approach

Inferring Relationship of Blood Metabolic Changes and Average Daily Gain With Feed Conversion Efficiency in Murrah Heifers: Machine Learning Approach

Preharvest phenotypic prediction of grain quality and yield of durum wheat using multispectral imaging

The spatial dynamics of wheat yield and protein content at the field scale

Contact Info

Product

Resources

About