Winter wheat is a widely-grown cereal crop worldwide. Using growth-stage information to estimate winter wheat yields in a timely manner is essential for accurate crop management and rapid decision-making in sustainable agriculture, and to increase productivity while reducing environmental impact. UAV remote sensing is widely used in precision agriculture due to its flexibility and increased spatial and spectral resolution. Hyperspectral data are used to model crop traits because of their ability to provide continuous rich spectral information and higher spectral fidelity. In this study, hyperspectral image data of the winter wheat crop canopy at the flowering and grain-filling stages was acquired by a low-altitude unmanned aerial vehicle (UAV), and machine learning was used to predict winter wheat yields. Specifically, a large number of spectral indices were extracted from the spectral data, and three feature selection methods, recursive feature elimination (RFE), Boruta feature selection, and the Pearson correlation coefficient (PCC), were used to filter high spectral indices in order to reduce the dimensionality of the data. Four major basic learner models, (1) support vector machine (SVM), (2) Gaussian process (GP), (3) linear ridge regression (LRR), and (4) random forest (RF), were also constructed, and an ensemble machine learning model was developed by combining the four base learner models. The results showed that the SVM yield prediction model, constructed on the basis of the preferred features, performed the best among the base learner models, with an R2 between 0.62 and 0.73. The accuracy of the proposed ensemble learner model was higher than that of each base learner model; moreover, the R2 (0.78) for the yield prediction model based on Boruta’s preferred characteristics was the highest at the grain-filling stage.
The objective of this study was to test the effects of N rates and tillage radish (Raphanus sativus var. longipinnatus) cover crop (TRCC) on soil water and cotton (Gossypium hirsutum L.) yield. In three years of the investigation, the treatments were N rates at 84 kg ha−1 and 140 kg ha−1 with and without TRCC. Soil water contents were measured using soil water sensors. Results showed that cotton yield was not significantly (p > 0.05) influenced by TRCC. Compared to N rate at 84 kg ha−1, 140 kg N ha−1 increased lint yield by 2.0%, 7.4%, 18.4% in 2017, 2018, and 2019, respectively, but the increase was significant only in 2019 (p < 0.02). Interactions between TRCC and nitrogen rate on yield were significant (p < 0.03) only in 2017. TRCC increased soil water infiltration capacity, resulting in higher soil water content. Use of TRCC did not affect the cotton yield, which could be due to the high inputs of water and high rates of N neutralizing the positive contributions to the cotton growth expected from the TRCC. Sub-optimum winter temperatures hampered the establishment and subsequent growth of TRCC, which also possibly contributed to its minimum impacts on cotton crop performance in the following season.
Saw-type lint cleaner (STLC) was most efficient lint cleaner in cotton ginning. However, STLC damaged fiber quality. An air-bar lint cleaner (ABLC) was developed and evaluated to preserve cotton fiber quality. The ABLC used pressurized-air to remove non-lint materials from cotton fiber. During lint cleaning process, non-lint materials attached to the fiber were blown off the fiber without the fiber making aggressive mechanical contact with a grid bar in conventional saw-type lint cleaner (STLC). It was expected using this concept that the fiber quality could be preserved by reducing the damage from mechanical impact of the fiber against the grid bar. Preliminary testing of the ABLC prototype showed that ABLC generated less lint waste and had a higher turnout rate than STLC. Use of ABLC could save 2.8 kg of lint in each 225 kg bale of cotton. The High Volume Instrument (HVI) analysis indicated the fiber properties in fiber length, uniformity, short fiber content, and color were not significantly different between ABLC and STLC. However, the Advanced Fiber Information System (AFIS) tests showed STLC had better performance than ABLC in fiber length and short fiber content while the trash and dust content with ABLC was lower than the STLC. More research was necessary to further prove the concept of ABLC and improve its performance in preserving cotton fiber quality.