Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Nov 9;22(22):8629.
doi: 10.3390/s22228629.

Performance and Accuracy Comparisons of Classification Methods and Perspective Solutions for UAV-Based Near-Real-Time "Out of the Lab" Data Processing

Zsófia Varga  1 , Fanni Vörös  1 , Márton Pál  1 , Béla Kovács  1 , András Jung  1 , István Elek  1
Affiliations

Performance and Accuracy Comparisons of Classification Methods and Perspective Solutions for UAV-Based Near-Real-Time "Out of the Lab" Data Processing

Zsófia Varga et al. Sensors (Basel). .

Abstract

Today, integration into automated systems has become a priority in the development of remote sensing sensors carried on drones. For this purpose, the primary task is to achieve real-time data processing. Increasing sensor resolution, fast data capture and the simultaneous use of multiple sensors is one direction of development. However, this poses challenges on the data processing side due to the increasing amount of data. Our study intends to investigate how the running time and accuracy of commonly used image classification algorithms evolve using Altum Micasense multispectral and thermal acquisition data with GSD = 2 cm spatial resolution. The running times were examined for two PC configurations, with a 4 GB and 8 GB DRAM capacity, respectively, as these parameters are closer to the memory of NRT microcomputers and laptops, which can be applied "out of the lab". During the accuracy assessment, we compared the accuracy %, the Kappa index value and the area ratio of correct pixels. According to our results, in the case of plant cover, the Spectral Angles Mapper (SAM) method achieved the best accuracy among the validated classification solutions. In contrast, the Minimum Distance (MD) method achieved the best accuracy on water surface. In terms of temporality, the best results were obtained with the individually constructed decision tree classification. Thus, it is worth developing these two directions into real-time data processing solutions.

Keywords: CTA analysis; GIS data NRT processing “out of the lab”; PCA analysis; UAV multispectral images classifications; classification’s running time; decision rules.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Study area on Lake Tisza, Hungary: (a) True color composite of Sentinel MSIL2A satellite image at the same time after preprocessing. (b) True color composite of the Micasense Altum 5 band stacked orthophoto after preprocessing (the centroid’s coordinates of the surveyed area: Lat = N47.659, Lon = E20.717, EPSG = 4326).
Figure 2
Figure 2
Reference surface vegetation cover based on field survey.
Figure 3
Figure 3
The schematic flowchart of our workflow with the examined algorithms.
Figure 4
Figure 4
Total area (m2) of the correct pixels by classes broken down by the used classification algorithm.
Figure 5
Figure 5
(a) Overall accuracy % of the classification algorithms by the classes and the overall model; (b) Kappa indexes of the classification algorithms by the classes and the overall model.
Figure 6
Figure 6
(a) Overall accuracy % of the classifications used four classes; (b) Kappa index of the classifications used four classes.
Figure 7
Figure 7
Images of the results of PCA analysis: (a) PCA level 1. (b) PCA level 2. (c) PCA level 3. (d) PCA level 4. (e) PCA level 5.
Figure 8
Figure 8
(a) The overall accuracy (%) of the PCA analysis and the reference layer. (b) The results of Kappa index of the PCA analysis and the reference layer.
Figure 9
Figure 9
(a) Overall accuracy % by the PCA and the UAV ‘original’ classifications. (b) Kappa indexes of PCA and the UAV ‘original’ classifications.
Figure 10
Figure 10
(a) Results of accuracy assessment of the decision rules for fairy rose. (b) Results of Kappa index of the decision rules for fairy rose.
See this image and copyright information in PMC

References

    1. Fan B., Li Y., Zhang R., Fu Q. Review on the Technological Development and Application of UAV Systems. Chin. J. Electron. 2020;29:199–207. doi: 10.1049/cje.2019.12.006. - DOI
    1. Tsouros D.C., Bibi S., Sarigiannidis P.G. A Review on UAV-Based Applications for Precision Agriculture. Information. 2019;10:349. doi: 10.3390/info10110349. - DOI
    1. Liu J., Xiang J., Jin Y., Liu R., Yan J., Wang L. Boost Precision Agriculture with Unmanned Aerial Vehicle Remote Sensing and Edge Intelligence: A Survey. Remote Sens. 2021;13:4387. doi: 10.3390/rs13214387. - DOI
    1. Jung A., Vohland M. Hyperspectral Remote Sensing and Field Spectroscopy: Applications in Agroecology and Organic Farming. In: De Marchi M., Diantini A., editors. Drones Information Technologies in Agroecology and Organic Farming Contributions to Technological Sovereignty. 1st ed. CRC Press; Boca Raton, FL, USA: 2022. pp. 99–121. - DOI
    1. Willkomm M., Bolten A., Bareth G. Non-destructive monitoring of rice by hyperspectral in-field spectrometry and UAV-based remote sensing: Case study of field-grown rice in north Rhine-Westphalia, Germany. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2016;XLI-B1:1071–1077. doi: 10.5194/isprs-archives-XLI-B1-1071-2016. - DOI

LinkOut - more resources