Deep Learning Neural Networks for Land Use Land Cover Mapping

This analysis determined the best individual band and combinations of various numbers of bands for land use land cover mapping for three sites in Peru. The data included Landsat Thematic Mapper (TM) optical data, PALSAR L-band dual-polarized radar, and derived radar texture images. Spectral signatures were first obtained for each site class and separability between classes determined using divergence measures. Results show that the best single band for analysis was a TM band, which was different for each site. For two of the three sites, the second best band was a radar texture image from a large window size. For all sites the best three bands included two TM bands and a radar texture image. The original PALSAR bands were of limited value. Finally upon further analysis it was determined that no more than six bands were needed for viable classification at each study site.

Download Full-text

Land Use/Land Cover Mapping from Airborne Hyperspectral Images with Machine Learning Algorithms and Contextual Information

Geocarto International ◽

10.1080/10106049.2021.1945149 ◽

2021 ◽

pp. 1-40

Author(s):

Ozlem Akar ◽

Esra Tunc Gormus

Keyword(s):

Machine Learning ◽

Land Use ◽

Land Cover ◽

Contextual Information ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Hyperspectral Images ◽

Land Cover Mapping ◽

Land Use Land Cover

Download Full-text

A Comparative Study of Drone and High Resolution Satellite Data for Detailed Land Use/Land Cover Mapping

Lecture Notes in Civil Engineering - Proceedings of UASG 2019 ◽

10.1007/978-3-030-37393-1_1 ◽

2020 ◽

pp. 1-10

Author(s):

Ajay Mathur ◽

P. K. Litoria ◽

B. Pateriya

Keyword(s):

Land Use ◽

High Resolution ◽

Land Cover ◽

Comparative Study ◽

Satellite Data ◽

Land Cover Mapping ◽

Land Use Land Cover

Download Full-text

Land Use/Land Cover Mapping of Sambar Watershed by Using Remote Sensing and GIS

International Journal of Current Microbiology and Applied Sciences ◽

10.20546/ijcmas.2021.1009.066 ◽

2021 ◽

Vol 10 (9) ◽

pp. 578-585

Author(s):

B. Varpe Shriniwas D. Payal Sandip

Keyword(s):

Remote Sensing ◽

Land Use ◽

Land Cover ◽

Remote Sensing And Gis ◽

Rapid Expansion ◽

Water Bodies ◽

Land Cover Mapping ◽

Fallow Land ◽

Land Use Land Cover ◽

Barren Land

In the present study, an effort has been made to study in detail of Land Use/Land Cover Mapping for Sambar watershed by using Remote Sensing and GIS technique was carried out during the year of 2020-2021 in Parbhani district. In this research the Remote Sensing and Geographical Information system technique was used for identifying the land use/land cover classes with the help of ArcGIS 10.8 software. The Sambar watershed is located in 19º35ʹ78.78˝ N and 76º87ʹ88.44˝ E in the Parbhani district of Marathwada region in Maharashtra. It is covered a total area 97.01 km2. The land use/land cover map and its classes were identified by the Supervised Classification Method in ArcGIS 10.8 software by using the Landsat 8 satellite image. Total six classes are identified namely as Agricultural area, Forest area, Urban area, Barren land, Water bodies and Fallow land. The Agricultural lands are well distributed throughout the watershed area and it covers 4135 ha. (43 per cent). Forest occupies 502 ha area and sharing about 5 per cent of the total land use land cover of the study area. The Urban land occupies 390 ha. area (4 per cent) and there was a rapid expansion of settlement area. Barren land occupies 3392 ha. area (35 per cent). A water bodies occupy 630 ha. area (6 per cent) and the Fallow land occupies 650 ha (7 per cent) but well-developed dendritic drainage pattern and good water availability is in the Sambar watershed.

Download Full-text