Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria
A classification method of hyperspectral reflectance images named CHRIPS (Classification of Hyperspectral Reflectance Images with Physical and Statistical criteria) is presented. This method aims at classifying each pixel from a given set of thirteen classes: unidentified dark surface, water, plasti...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-07-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/12/14/2335 |
_version_ | 1797561797433098240 |
---|---|
author | Alexandre Alakian Véronique Achard |
author_facet | Alexandre Alakian Véronique Achard |
author_sort | Alexandre Alakian |
collection | DOAJ |
description | A classification method of hyperspectral reflectance images named CHRIPS (Classification of Hyperspectral Reflectance Images with Physical and Statistical criteria) is presented. This method aims at classifying each pixel from a given set of thirteen classes: unidentified dark surface, water, plastic matter, carbonate, clay, vegetation (dark green, dense green, sparse green, stressed), house roof/tile, asphalt, vehicle/paint/metal surface and non-carbonated gravel. Each class is characterized by physical criteria (detection of specific absorptions or shape features) or statistical criteria (use of dedicated spectral indices) over spectral reflectance. CHRIPS input is a hyperspectral reflectance image covering the spectral range [400–2500 nm]. The presented method has four advantages, namely: (i) is robust in transfer, class identification is based on criteria that are not very sensitive to sensor type; (ii) does not require training, criteria are pre-defined; (iii) includes a reject class, this class reduces misclassifications; (iv) high precision and recall, F<inline-formula> <math display="inline"> <semantics> <msub> <mrow></mrow> <mn>1</mn> </msub> </semantics> </math> </inline-formula> score is generally above 0.9 in our test. As the number of classes is limited, CHRIPS could be used in combination with other classification algorithms able to process the reject class in order to decrease the number of unclassified pixels. |
first_indexed | 2024-03-10T18:19:44Z |
format | Article |
id | doaj.art-4a65d79bb70844ed85ae21273b6a00f1 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T18:19:44Z |
publishDate | 2020-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-4a65d79bb70844ed85ae21273b6a00f12023-11-20T07:25:04ZengMDPI AGRemote Sensing2072-42922020-07-011214233510.3390/rs12142335Classification of Hyperspectral Reflectance Images With Physical and Statistical CriteriaAlexandre Alakian0Véronique Achard1ONERA, The French Aerospace Lab, Université Paris Saclay, FR-91123 Palaiseau, FranceONERA, The French Aerospace Lab, 2 Avenue Edouard Belin, CEDEX, 31055 Toulouse, FranceA classification method of hyperspectral reflectance images named CHRIPS (Classification of Hyperspectral Reflectance Images with Physical and Statistical criteria) is presented. This method aims at classifying each pixel from a given set of thirteen classes: unidentified dark surface, water, plastic matter, carbonate, clay, vegetation (dark green, dense green, sparse green, stressed), house roof/tile, asphalt, vehicle/paint/metal surface and non-carbonated gravel. Each class is characterized by physical criteria (detection of specific absorptions or shape features) or statistical criteria (use of dedicated spectral indices) over spectral reflectance. CHRIPS input is a hyperspectral reflectance image covering the spectral range [400–2500 nm]. The presented method has four advantages, namely: (i) is robust in transfer, class identification is based on criteria that are not very sensitive to sensor type; (ii) does not require training, criteria are pre-defined; (iii) includes a reject class, this class reduces misclassifications; (iv) high precision and recall, F<inline-formula> <math display="inline"> <semantics> <msub> <mrow></mrow> <mn>1</mn> </msub> </semantics> </math> </inline-formula> score is generally above 0.9 in our test. As the number of classes is limited, CHRIPS could be used in combination with other classification algorithms able to process the reject class in order to decrease the number of unclassified pixels.https://www.mdpi.com/2072-4292/12/14/2335classificationhyperspectralreflectancereject classspecific absorptionspectral index |
spellingShingle | Alexandre Alakian Véronique Achard Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria Remote Sensing classification hyperspectral reflectance reject class specific absorption spectral index |
title | Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria |
title_full | Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria |
title_fullStr | Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria |
title_full_unstemmed | Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria |
title_short | Classification of Hyperspectral Reflectance Images With Physical and Statistical Criteria |
title_sort | classification of hyperspectral reflectance images with physical and statistical criteria |
topic | classification hyperspectral reflectance reject class specific absorption spectral index |
url | https://www.mdpi.com/2072-4292/12/14/2335 |
work_keys_str_mv | AT alexandrealakian classificationofhyperspectralreflectanceimageswithphysicalandstatisticalcriteria AT veroniqueachard classificationofhyperspectralreflectanceimageswithphysicalandstatisticalcriteria |