| Title | Lapų atspindžio spektrų požymių atrankos tyrimas automatiniam trąšų poreikio nustatymui |
|---|---|
| Translation of Title | Feature selection analysis of crop leaf reflectance spectra for automated nutrient deficiency detection. |
| Authors | Mykolaitis, Paulius |
| Full Text |
|
| Pages | 49 |
| Keywords [eng] | support vector machines ; nearest neighbor algorithm ; feature selection ; reflection spectroscopy ; crops nutrient deficiency |
| Abstract [eng] | A literature overview of plant nutrient deficiency diagnostic methods, classification algorithms (k-NN, SVM, Random Forest), and feature selection methods (Laplace, Fisher, ReliefF, χ²) are provided in the thesis. The main objective of the study is to compare different feature selection methods within the context of Support Vector Machine (SVM) classification algorithm and identify the most important features that can be further analyzed by domain experts in order to better understand the plant physiological processes in plants with nutrient deficiencies. |
| Dissertation Institution | Kauno technologijos universitetas. |
| Type | Master thesis |
| Language | Lithuanian |
| Publication date | 2024 |