Переведення RGB зображень у гіперспектральні за допомогою алгоритмів машинного навчання
Loading...
Date
2023
Authors
Устименко, Данило
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This work reviews the problem of reconstruction of hyperspectral images from RGB ones. The solution is proposed with a U-net model. Some experiments on the improvement of the performance are made. The model is used for the real case classification problem. Even though it did not improve the outcome of the second problem, the proper analysis of the problems faced and future solutions was made.
Description
Keywords
hyperspectral images, convolutional neural networks, U-net, EfficientNet, spectre, data augmentation, курсова робота