Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence

Cidália C. Fonte, Luísa M. S. Gonçalves

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Downloads (Pure)

Abstract

The aim of this article is to assess if the data provided by soft classifiers and uncertainty measures can be used to identify regions with different levels of accuracy in a classified image. To this aim a soft Bayesian classifier was used, which enables the assignment of classifications confidence levels to all pixels. Two uncertainty measures were also used, namely the Relative Maximum Deviation (RMD) uncertainty measure and the Normalized Entropy (NE). The approach was tested on a case study. A multispectral IKONOS image was classified and the classification uncertainty and confidence where computed and analysed. Regions with different levels of uncertainty and confidence were identified. Reference datasets were then used to assess the classification accuracy of the whole study area and also in the regions with different levels of uncertainty and confidence. A comparative analysis was made on the variation of accuracy and classification uncertainty and confidence along the map and per class. The results show that for the regions with more uncertainty or less confidence the spatially constrained confusion matrices always generate lower values of global accuracy than for global accuracy of the regions with less uncertainty or more confidence. The analysis of the user’s and producer’s accuracy also shows the same general tendency. Proposals are then made on methodologies to use the information provided by the uncertainty and confidence to identify less reliable regions and also to improve classification results using fully automated approaches.

Original languageEnglish
Title of host publicationSPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change”
Pages275-281
Number of pages7
Volume42
Edition4
DOIs
Publication statusPublished - 19 Sep 2018
EventISPRS TC IV Mid-Term Symposium on 3D Spatial Information Science - The Engine of Change - Delft, Netherlands
Duration: 1 Oct 20185 Oct 2018

Publication series

NameInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
ISSN (Print)1682-1750

Conference

ConferenceISPRS TC IV Mid-Term Symposium on 3D Spatial Information Science - The Engine of Change
CountryNetherlands
CityDelft
Period1/10/185/10/18

Fingerprint

land cover
confidence
uncertainty
IKONOS
Classifiers
entropy
Uncertainty
pixel
matrix
methodology
producer
Entropy
Pixels
analysis

Keywords

  • Accuracy
  • Classification
  • Confidence
  • Multispectral images
  • Spatial variation
  • Uncertainty

Cite this

Fonte, C. C., & Gonçalves, L. M. S. (2018). Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. In SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” (4 ed., Vol. 42, pp. 275-281). (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives). https://doi.org/10.5194/isprs-archives-XLII-4-201-2018
Fonte, Cidália C. ; Gonçalves, Luísa M. S. . / Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” . Vol. 42 4. ed. 2018. pp. 275-281 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives).
@inproceedings{18c2d727d3fa4d6991477c5db439e42e,
title = "Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence",
abstract = "The aim of this article is to assess if the data provided by soft classifiers and uncertainty measures can be used to identify regions with different levels of accuracy in a classified image. To this aim a soft Bayesian classifier was used, which enables the assignment of classifications confidence levels to all pixels. Two uncertainty measures were also used, namely the Relative Maximum Deviation (RMD) uncertainty measure and the Normalized Entropy (NE). The approach was tested on a case study. A multispectral IKONOS image was classified and the classification uncertainty and confidence where computed and analysed. Regions with different levels of uncertainty and confidence were identified. Reference datasets were then used to assess the classification accuracy of the whole study area and also in the regions with different levels of uncertainty and confidence. A comparative analysis was made on the variation of accuracy and classification uncertainty and confidence along the map and per class. The results show that for the regions with more uncertainty or less confidence the spatially constrained confusion matrices always generate lower values of global accuracy than for global accuracy of the regions with less uncertainty or more confidence. The analysis of the user’s and producer’s accuracy also shows the same general tendency. Proposals are then made on methodologies to use the information provided by the uncertainty and confidence to identify less reliable regions and also to improve classification results using fully automated approaches.",
keywords = "Accuracy, Classification, Confidence, Multispectral images, Spatial variation, Uncertainty",
author = "Fonte, {Cid{\'a}lia C.} and Gon{\cc}alves, {Lu{\'i}sa M. S.}",
note = "Fonte, C. C., & Gon{\cc}alves, L. M. S. (2018). Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. In SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” (4 ed., Vol. 42, pp. 275-281). (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives). DOI: 10.5194/isprs-archives-XLII-4-201-2018",
year = "2018",
month = "9",
day = "19",
doi = "10.5194/isprs-archives-XLII-4-201-2018",
language = "English",
volume = "42",
series = "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives",
pages = "275--281",
booktitle = "SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change”",
edition = "4",

}

Fonte, CC & Gonçalves, LMS 2018, Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. in SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” . 4 edn, vol. 42, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, pp. 275-281, ISPRS TC IV Mid-Term Symposium on 3D Spatial Information Science - The Engine of Change, Delft, Netherlands, 1/10/18. https://doi.org/10.5194/isprs-archives-XLII-4-201-2018

Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. / Fonte, Cidália C.; Gonçalves, Luísa M. S. .

SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” . Vol. 42 4. ed. 2018. p. 275-281 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence

AU - Fonte, Cidália C.

AU - Gonçalves, Luísa M. S.

N1 - Fonte, C. C., & Gonçalves, L. M. S. (2018). Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. In SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” (4 ed., Vol. 42, pp. 275-281). (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives). DOI: 10.5194/isprs-archives-XLII-4-201-2018

PY - 2018/9/19

Y1 - 2018/9/19

N2 - The aim of this article is to assess if the data provided by soft classifiers and uncertainty measures can be used to identify regions with different levels of accuracy in a classified image. To this aim a soft Bayesian classifier was used, which enables the assignment of classifications confidence levels to all pixels. Two uncertainty measures were also used, namely the Relative Maximum Deviation (RMD) uncertainty measure and the Normalized Entropy (NE). The approach was tested on a case study. A multispectral IKONOS image was classified and the classification uncertainty and confidence where computed and analysed. Regions with different levels of uncertainty and confidence were identified. Reference datasets were then used to assess the classification accuracy of the whole study area and also in the regions with different levels of uncertainty and confidence. A comparative analysis was made on the variation of accuracy and classification uncertainty and confidence along the map and per class. The results show that for the regions with more uncertainty or less confidence the spatially constrained confusion matrices always generate lower values of global accuracy than for global accuracy of the regions with less uncertainty or more confidence. The analysis of the user’s and producer’s accuracy also shows the same general tendency. Proposals are then made on methodologies to use the information provided by the uncertainty and confidence to identify less reliable regions and also to improve classification results using fully automated approaches.

AB - The aim of this article is to assess if the data provided by soft classifiers and uncertainty measures can be used to identify regions with different levels of accuracy in a classified image. To this aim a soft Bayesian classifier was used, which enables the assignment of classifications confidence levels to all pixels. Two uncertainty measures were also used, namely the Relative Maximum Deviation (RMD) uncertainty measure and the Normalized Entropy (NE). The approach was tested on a case study. A multispectral IKONOS image was classified and the classification uncertainty and confidence where computed and analysed. Regions with different levels of uncertainty and confidence were identified. Reference datasets were then used to assess the classification accuracy of the whole study area and also in the regions with different levels of uncertainty and confidence. A comparative analysis was made on the variation of accuracy and classification uncertainty and confidence along the map and per class. The results show that for the regions with more uncertainty or less confidence the spatially constrained confusion matrices always generate lower values of global accuracy than for global accuracy of the regions with less uncertainty or more confidence. The analysis of the user’s and producer’s accuracy also shows the same general tendency. Proposals are then made on methodologies to use the information provided by the uncertainty and confidence to identify less reliable regions and also to improve classification results using fully automated approaches.

KW - Accuracy

KW - Classification

KW - Confidence

KW - Multispectral images

KW - Spatial variation

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=85056208772&partnerID=8YFLogxK

U2 - 10.5194/isprs-archives-XLII-4-201-2018

DO - 10.5194/isprs-archives-XLII-4-201-2018

M3 - Conference contribution

VL - 42

T3 - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

SP - 275

EP - 281

BT - SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change”

ER -

Fonte CC, Gonçalves LMS. Identification of low accuracy regions in land cover maps using uncertainty measures and classification confidence. In SPRS TC IV Mid-term Symposium “3D Spatial Information Science – The Engine of Change” . 4 ed. Vol. 42. 2018. p. 275-281. (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives). https://doi.org/10.5194/isprs-archives-XLII-4-201-2018