Image Simplification Using Kohonen Maps: Application to Satellite Data for Cloud Detection and Land Cover Mapping

Land cover mapping of large areas is challenging due to the significant volume of satellite data to acquire and process, as well as the lack of spatial continuity due to cloud cover. Temporal aggregation—the use of metrics (i.e., mean or median) derived from satellite data over a period of time—is an approach that benefits from recent increases in the frequency of free satellite data acquisition and cloud-computing power. This enables the efficient use of multi-temporal data and the exploitation of cloud-gap filling techniques for land cover mapping. Here, we provide the first formal comparison of the accuracy between land cover maps created with temporal aggregation of Sentinel-1 (S1), Sentinel-2 (S2), and Landsat-8 (L8) data from one-year and test whether this method matches the accuracy of traditional approaches. Thirty-two datasets were created for Wales by applying automated cloud-masking and temporally aggregating data over different time intervals, using Google Earth Engine. Manually processed S2 data was used for comparison using a traditional two-date composite approach. Supervised classifications were created, and their accuracy was assessed using field-based data. Temporal aggregation only matched the accuracy of the traditional two-date composite approach (77.9%) when an optimal combination of optical and radar data was used (76.5%). Combined datasets (S1, S2 or S1, S2, and L8) outperformed single-sensor datasets, while datasets based on spectral indices obtained the lowest levels of accuracy. The analysis of cloud cover showed that to ensure at least one cloud-free pixel per time interval, a maximum of two intervals per year for temporal aggregation were possible with L8, while three or four intervals could be used for S2. This study demonstrates that temporal aggregation is a promising tool for integrating large amounts of data in an efficient way and that it can compensate for the lower quality of automatic image selection and cloud masking. It also shows that combining data from different sensors can improve classification accuracy. However, this study highlights the need for identifying optimal combinations of satellite data and aggregation parameters in order to match the accuracy of manually selected and processed image composites.

Download Full-text

Land use — land cover mapping using satellite data for a forested watershed, Udupi district, Karnataka state, India

Journal of the Indian Society of Remote Sensing ◽

10.1007/bf02990040 ◽

2005 ◽

Vol 33 (2) ◽

pp. 233-238 ◽

Cited By ~ 8

Author(s):

Amba Shetty ◽

Lakshman Nandagiri ◽

Sangeeta Thokchom ◽

M. V. S. Rajesh

Keyword(s):

Land Use ◽

Land Cover ◽

Satellite Data ◽

Forested Watershed ◽

Land Cover Mapping ◽

Land Use Land Cover ◽

Karnataka State

Download Full-text

Assessment of land-use and land-cover changes in Pangari watershed area (MS), India, based on the remote sensing and GIS techniques

Applied Water Science ◽

10.1007/s13201-021-01425-1 ◽

2021 ◽

Vol 11 (6) ◽

Author(s):

Chaitanya B. Pande ◽

Kanak N. Moharir ◽

S. F. R. Khadri

Keyword(s):

Remote Sensing ◽

Land Use ◽

Land Cover ◽

Satellite Data ◽

Supervised Classification ◽

Remote Sensing And Gis ◽

Land Use Policy ◽

Classification Method ◽

Land Cover Mapping ◽

Planning And Management

AbstractIn this paper, we focus on the assessment of land-use and land-cover change detection mapping to the effective planning and management policies of environment, land-use policy and hydrological system in the study area. In this study the soil and water conservation project has been applied during the five years and after five years what changes have been found in the land-use and land-cover classes and vegetation. In this view, this land-use and land-cover mapping is a more important role to decide the policy for watershed planning and management project in the semiarid region. In an emerging countries, fast industrialization and urbanization impose a significant threat to the natural atmosphere. The remote sensing and GIS techniques are crucial roles in the study of land-use and land-cover mapping during the years of 2007, 2014, and 2017. The main objective of this is to prepare the land-use and NDVI maps in the years of 2008, 2014 and 2017; these maps have prepared from satellite data using the supervised classification method. A normalized difference vegetation index map (NDVI) was done by using Landsat 8 and LISS-III satellite data. NDVI values play a major role in monitoring the vegetation and variation in land-use and land-cover classes. In these maps, four types of land are divided into four classes as agriculture, built-up, wasteland, and water body. The results of study show that agriculture land of 18.71% (158.24 Ha), built-up land of 0.62% (5.31 Ha), wasteland of 40.33% (341.02 Ha), and water body land of 17.39% (147 Ha) are increased. Land-use and land-cover maps and NDVI values show that agriculture land of 22.97% (194.29 Ha), 5.46% (14.59 Ha), and 0.08% (0.22 Ha) decreases during the years of 2008, 2014, and 2017. The results directly indicate that the supervised classification method has been the accurate identified feature in the land-use map classes. This classification method has been given the better accuracy (95%) from spatiotemporal satellite data. The accuracy was also tally with ground-truth and Google earth information. These results can be a very useful for the land-use policy, watershed planning, and management with natural resources, animals, and ecological systems.

Download Full-text