Filters
Application of functional analysis in dendrometry using five-year growth of selected dendrometric traits of scots pine Pinus sylvestris L.)
Abstract The differentiation between age classes of Scots pine (Pinus sylvestris L.) was analyzed with regard to the five-year increment of seven traits: height growth (zh5), diameter growth at breast height (zd5), cross-sectional area growth at breast height (zg5), volume growth (zv5), volume growth intensity coefficient (i5), and slenderness (s). Measurements were made in five periods for 24-year-old trees and six periods for 33-year-old trees, all growing in fresh mixed coniferous forest sites. Repeated measures data analysis was conducted separately for all traits. Multivariable functional data analysis (FDA) was proposed to compare age classes of trees. The functional variables which resulted from this analysis can be used, as data, in many analyses (designate functions representing each of trees, FPCA – functional principal component analysis, FLDC – discriminant analysis, permutation analysis of variance). The results of the above analyses revealed significant differences between age groups. Furthermore the functions and FPCA were used to detect outliers. This procedure had not previously been used for such a purpose. FPCA explained 55% of the total variance, with the first two components clearly separating the groups. The study showed that 33-year-old trees exhibit stable growth, while 24-year-old trees show greater variability, highlighting the impact of age on growth dynamics. Permutation analysis of variance confirmed significant growth differences between the groups. The findings highlight the importance of age as a factor influencing tree growth and demonstrate the effectiveness of the multivariable FDA approach for analyzing such data.
Environmental effects on physiological index of black alder (Alnus glutinosa [L.] Gaertn.) dominant trees in central Bosnia
Progress in plant tolerance to the fungal disease Sclerotinia in breeding experiments on winter oilseed rape
Summary Sclerotinia sclerotiorum is a pathogen which causes a disease of oilseed rape. Severe plant infection contributes to a decrease in the quality of the crop. It is therefore important to pay attention to whether hybrids are highly tolerant to this fungal disease at the early stages of breeding. In this study, the question of interest is whether there has been progress in increasing the tolerance of new hybrids to this pathogen. Three years of breeding experiments (2014, 2015, 2017) are included in the analysis. Each year, three to five experiments were carried out, with several dozen varieties and three standards. Each series of experiments was repeated in several locations. Because the degree of infection was assessed on a scale (from 1 – the highest infection – to 9 – the least infection), the analysis is carried out using an ordinal logistic model. It is noted that in earlier years the standard varieties’ probability of infection with this disease had a smaller range (empirical probability 0.5–0.83) than in the last analyzed year (empirical probability 0.33–1.00). The results of the analysis show that in 2014 and 2015 several hybrids exhibited a significantly higher tolerance to Sclerotinia, but in 2017 none of the hybrids were significantly better than the standard. Perhaps breeding selection of hybrids has eliminated the less tolerant varieties. However, to be able to draw more general conclusions, it would be necessary to repeat the study in controlled conditions (a greenhouse), where the level of fungal spores and their effect on plants could be controlled. Obtaining tolerant hybrids will enable a reduction in production costs, since there will be no need to monitor whether disease infestation occurs and no need to use corresponding plant protection products.
Applying convolutional neural networks for mustard variety recognition
Abstract The aim of this study was to develop and apply a Convolutional Neural Network (CNN) model to recognize and classify white mustard (Sinapis Alba L.) varieties, addressing the complex task of discriminating among 57 varieties. Utilizing a one-dimensional CNN model, the research focused on multivariate analysis based on a set of 15 traits. The CNN architecture included convolutional layers, batch normalization, pooling, flattening, dropout, and dense layers. The model demonstrated effectiveness in classifying varieties, achieving high accuracy and providing valuable insights into potential new varieties. Subset division, a new approach, was applied. Evaluation metrics, including accuracy, F1 score, precision, and recall, were calculated for eight subsets, confirming the model's robust performance. While this study uses mustard as an illustrative example, the method is not limited to this crop and can be extended to other agricultural crops, with potential modifications depending on the specific traits relevant to each crop. The approach contributes to agricultural advancements, offering a reliable tool for breeders to assess variety distinctness and streamline the testing process. The model’s ability to detect unknown varieties further enhances its utility in agricultural research covering a comprehensive and impactful advancement in variety classification.