Spectral-Spatial Methods for Hyperspectral Image Classification. Review

S. M. Borzov; O. I. Potaturkin

doi:10.3103/S8756699018060079

Spectral-Spatial Methods for Hyperspectral Image Classification. Review

S. M. Borzov, O. I. Potaturkin

Результат исследования: Научные публикации в периодических изданиях › статья

3 Цитирования (Scopus)

Аннотация

Various methods of spectral-spatial classification of hyperspectral data are reviewed. Papers devoted to the most popular ways of using spatial information for increasing the accuracy of classification maps are considered. It is shown that the best results are obtained by using preprocessing of “raw” data before the procedures of pixel-wise spectral classification. Disadvantages, limits, and possible directions for developing existing methods are investigated and analyzed.

Язык оригинала	английский
Страницы (с-по)	582-599
Число страниц	18
Журнал	Optoelectronics, Instrumentation and Data Processing
Том	54
Номер выпуска	6
DOI	https://doi.org/10.3103/S8756699018060079
Состояние	Опубликовано - 1 ноя 2018
Опубликовано для внешнего пользования	Да

Доступ к документу

10.3103/S8756699018060079

Link to publication in Scopus

Fingerprint Подробные сведения о темах исследования «Spectral-Spatial Methods for Hyperspectral Image Classification. Review». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать

@article{df9134f11dff4b9ba0a99b016f00d5cf,

title = "Spectral-Spatial Methods for Hyperspectral Image Classification. Review",

abstract = "Various methods of spectral-spatial classification of hyperspectral data are reviewed. Papers devoted to the most popular ways of using spatial information for increasing the accuracy of classification maps are considered. It is shown that the best results are obtained by using preprocessing of “raw” data before the procedures of pixel-wise spectral classification. Disadvantages, limits, and possible directions for developing existing methods are investigated and analyzed.",

keywords = "hyperspectral images, remote sensing, spectral and spatial features, surface type classification, ATTRIBUTE PROFILES, SEGMENTATION, SVMS",

author = "Borzov, {S. M.} and Potaturkin, {O. I.}",

year = "2018",

month = nov,

day = "1",

doi = "10.3103/S8756699018060079",

language = "English",

volume = "54",

pages = "582--599",

journal = "Optoelectronics, Instrumentation and Data Processing",

issn = "8756-6990",

publisher = "Allerton Press Inc.",

number = "6",

}

TY - JOUR

T1 - Spectral-Spatial Methods for Hyperspectral Image Classification. Review

AU - Borzov, S. M.

AU - Potaturkin, O. I.

PY - 2018/11/1

Y1 - 2018/11/1

N2 - Various methods of spectral-spatial classification of hyperspectral data are reviewed. Papers devoted to the most popular ways of using spatial information for increasing the accuracy of classification maps are considered. It is shown that the best results are obtained by using preprocessing of “raw” data before the procedures of pixel-wise spectral classification. Disadvantages, limits, and possible directions for developing existing methods are investigated and analyzed.

AB - Various methods of spectral-spatial classification of hyperspectral data are reviewed. Papers devoted to the most popular ways of using spatial information for increasing the accuracy of classification maps are considered. It is shown that the best results are obtained by using preprocessing of “raw” data before the procedures of pixel-wise spectral classification. Disadvantages, limits, and possible directions for developing existing methods are investigated and analyzed.

KW - hyperspectral images

KW - remote sensing

KW - spectral and spatial features

KW - surface type classification

KW - ATTRIBUTE PROFILES

KW - SEGMENTATION

KW - SVMS

UR - http://www.scopus.com/inward/record.url?scp=85060816191&partnerID=8YFLogxK

U2 - 10.3103/S8756699018060079

DO - 10.3103/S8756699018060079

M3 - Article

AN - SCOPUS:85060816191

VL - 54

SP - 582

EP - 599

JO - Optoelectronics, Instrumentation and Data Processing

JF - Optoelectronics, Instrumentation and Data Processing

SN - 8756-6990

IS - 6

ER -