Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Hyperspectral Unmixing of Agricultural Images taken from UAV Using Adapted U-Net Architecture

Vytautas Paura, Virginijus Marcinkevičius

TL;DR

This paper creates a hyperspectral unmixing dataset, created from blueberry field data gathered by a hyperspectral camera mounted on a UAV and proposes a hyperspectral unmixing algorithm based on U-Net network architecture to achieve more accurate unmixing results on existing and newly created hyperspectral unmixing datasets.

Abstract

The hyperspectral unmixing method is an algorithm that extracts material (usually called endmember) data from hyperspectral data cube pixels along with their abundances. Due to a lower spatial resolution of hyperspectral sensors data in each of the pixels may contain mixed information from multiple endmembers. In this paper we create a hyperspectral unmixing dataset, created from blueberry field data gathered by a hyperspectral camera mounted on a UAV. We also propose a hyperspectral unmixing algorithm based on U-Net network architecture to achieve more accurate unmixing results on existing and newly created hyperspectral unmixing datasets.

Hyperspectral Unmixing of Agricultural Images taken from UAV Using Adapted U-Net Architecture

TL;DR

This paper creates a hyperspectral unmixing dataset, created from blueberry field data gathered by a hyperspectral camera mounted on a UAV and proposes a hyperspectral unmixing algorithm based on U-Net network architecture to achieve more accurate unmixing results on existing and newly created hyperspectral unmixing datasets.

Abstract

The hyperspectral unmixing method is an algorithm that extracts material (usually called endmember) data from hyperspectral data cube pixels along with their abundances. Due to a lower spatial resolution of hyperspectral sensors data in each of the pixels may contain mixed information from multiple endmembers. In this paper we create a hyperspectral unmixing dataset, created from blueberry field data gathered by a hyperspectral camera mounted on a UAV. We also propose a hyperspectral unmixing algorithm based on U-Net network architecture to achieve more accurate unmixing results on existing and newly created hyperspectral unmixing datasets.
Paper Structure (16 sections, 4 equations, 7 figures, 1 table)

This paper contains 16 sections, 4 equations, 7 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Hyperspectral unmixing algorithms
  4. Agricultural hyperspectral data unmixing
  5. Hyperspectral dataset of blueberry fields
  6. Data gathering
  7. Calibration
  8. Raw hyperspectral data
  9. Hyperspectral dataset ground truth creation
  10. Ground truth data mixing for hyperspectral unmixing dataset creation
  11. Hyperspectral unmixing using U-Net based architecture
  12. Metrics
  13. Proposed model architecture
  14. Datasets used for experimentation
  15. Evaluation of proposed method
  16. ...and 1 more sections

Figures (7)

  • Figure 1: Model predicted endmember (blue) comparison to ground truth endmember averages (orange) and their variations (green) for each of the six classes.
  • Figure 2: RGB representation of the hyperspectral blueberry data cubes, from top to bottom cubes 1, 2 and 3 are shown.
  • Figure 3: Averages, for each of the six classes, of extracted endmembers used as the ground truth for the created hyperspectral unmixing dataset.
  • Figure 4: Class distribution in the hyperspectral Cube 1. Color class representation: Yellow - bare soil; Green - Blueberries; Blue - Grass; Dark blue - Shadowed data; Light green - Water and wet soil; Black - Other data
  • Figure 5: Mixed hyperspectral cube RGB representation.
  • ...and 2 more figures