The classification of sugarcane varieties using products derived from remote sensing allows for the monitoring of plants with different profiles without necessarily having physical contact with the study objects. However, differentiating between varieties can be challenging due to the similarity of the spectral characteristics of each crop. Thus, this study aimed to classify four sugarcane varieties through deep neural networks, subsequently comparing the results with traditional machine learning techniques. In order to provide more data as input for the classification models, along with the multi-band values of the pixels and vegetation indices, other information can be obtained from the sensor bands through RGB combinations by reconciling different bands so as to yield the characteristics of crop varieties. The methodology created to discriminate sugarcane varieties consisted of a dense neural network, with the number of hidden layers determined by the greedy layer-wise method and multiples of four neurons in each layer; additionally, a 5-fold evaluation in the training data was composed of Sentinel-2 band data, vegetation indices, and RGB combinations. Comparing the results acquired from each model with the hyperparameters selected by Bayesian optimisation, except for the neural network with manually defined parameters, it was possible to observe a greater precision of 99.55% in the SVM model, followed by the neural network developed by the study, random forests, and kNN. However, the final neural network model prediction resulted in the 99.48% accuracy of a six-hidden-layers network, demonstrating the potential of using neural networks in classification. Among the characteristics that contributed the most to the classification, the chlorophyll-sensitive bands, especially B6, B7, B11, and some RGB combinations, had the most impact on the correct classification of samples by the neural network model. Thus, the regions encompassing the near-infrared and shortwave infrared regions proved to be suitable for the discrimination of sugarcane varieties.
No abstract
Testers for selection of sweet corn inbred lines Sweet corn is a special type of corn, with high nutritional value, which accumulates soluble polysaccharides in the endosperm, so it is sweeter than other types of corn. The consumption of sweet corn is increasing in Brazil. Being one of the major producers of corn in the world, there is also an enormous potential for the production of sweet corn. Currently, there are 53 sweet corn varieties registered in the country, but only one prevails in crops of this species. There is a demand for new varieties adapted to tropical conditions with high grain yield and superior grain quality. Additionally, there is not enough information about evaluation and development of sweet corn varieties for the Brazilian conditions. The goals of this project were: (i) to verify the feasibility of using a tester with high level of inbreeding and more selected in relation to testers with lower levels of inbreeding and less selected to obtain testcrosses; (ii) to verify the improvement of the testcross means; (iii) to estimate correlations between ear yield and production components; (iv) to apply a selection index to rank and select the best testcrosses and (v) to verify the genetic variance available after selection and selfing in the testcrosses. Two populations, where one was the tester of the other, were evaluated. In each cycle of selection testers were obtained from a mixture of selected inbred lines of the previous generation, leading to two levels of selection and three levels of inbreeding. For selection of inbred lines, 176 testcrosses were evaluated in two seasons, in randomized completed blocks with three replications. The traits days to male flowering (FM), plant height (AP), number of commercial ears (EC), ear yield (PE), ear length (CE), ear diameter (DE) and tip fill (EP) were evaluated. There were significant differences between testcross in the individual and the combined analysis of variance. Interactions between testcrosses and testers, from the individual analysis, in general, were not significant, indicating no change in the ranking of testcrosses when using different testers. For this interaction, from the joint analysis, Spearman's rank correlations were estimated and showed significance in most cases. For all variables, the testcross means were either equal to or greater than the best control. The level of selection underlying the constitution of a tester did not affect the genetic variance among testcrosses. The tester stemming from more selected inbred lines was as good as testers derived from less selected lines. There was progress in the testcross means, generally higher than 1%; the use of selected testers maximized testcross means. Positive genetic correlations (rG ≥ 0.556) were detected between ear yield and EC, CE, DE and EP traits. The selection index ranked the testcrosses for all traits evaluated simultaneously and allowed selecting the superior ones.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.