Accuracy assessment of land cover/land use classifiers in dry and humid areas of Iran

Yousefi, Saleh; Khatami, Reza; Mountrakis, Giorgos; Mirzaee, Somayeh; Pourghasemi, Hamid Reza; Tazeh, Mehdi

doi:10.1007/s10661-015-4847-1

Cited by 36 publications

(16 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main advantage for land cover mapping is the production of more accurate classification. (46)(47)(48) In our study also, the performance is better. MahD, which is a direction-sensitive distance classifier that uses statistics for each class and assumes that all class covariances are equal and therefore is a faster method, has also shown better performance and improvement.…”

Section: Resultssupporting

confidence: 62%

Exploring Land Cover Classification Accuracy of Landsat 8 Image Using Spectral Index Layer Stacking in Hilly Region of South Korea

Lee¹,

Acharya²,

Lee³

2018

Sensors and Materials

View full text Add to dashboard Cite

show abstract

Section: Resultssupporting

confidence: 62%

Exploring Land Cover Classification Accuracy of Landsat 8 Image Using Spectral Index Layer Stacking in Hilly Region of South Korea

Lee¹,

Acharya²,

Lee³

2018

Sensors and Materials

View full text Add to dashboard Cite

show abstract

“…The extracted training samples were divided randomly to 70% and 30% datasets, which were utilized for modeling and validation of LULC maps, respectively. Subsequently, a Support Vector Machine (SVM) algorithm was implemented for LULC mapping in the reference years [27]. Finally, Kappa coefficient and overall accuracy were computed and employed to assess the LULC maps through a comparison of produced LULC maps and real ground data [24,[27][28][29].…”

Section: Lulc Mapping and Detectionmentioning

confidence: 99%

Landscape Management through Change Processes Monitoring in Iran

et al. 2020

View full text Add to dashboard Cite

The presented research investigated and predicted landscape change processes (LCPs) in the Talar watershed, northern Iran. The Land Change Modeler was used for change analysis, transition potential modeling, and prediction of land use/land cover (LULC) map. The evaluation of projected LULC map was performed by comparing the real and predicted LULC maps for the reference year, 2014. Landscape metrics and change processes were investigated for the period 1989–2014 and for exploring the situation in 2030. Results illustrated that the increase in agricultural land and residential areas took place at the expense of forest and rangeland. The distance from forests was the most sensitive parameter for modeling the transition potentials. The modelling of the LULC change projected the number of patches, the landscape shape index, interspersion and juxtaposition index, and edge density, Euclidean nearest-neighbor distance, and area-weighted shape index will amount to 65.3, 7.63, 20.1, 8.77, −1.35, and 0.61% as compared to 2014, respectively. Our findings indicated that the type of change processes that occurred was not entirely the same in 1989–2000 and 2000–2014. In addition, change processes in the creation of dry farming, orchard, and residential classes, attrition of forest and rangeland categories, and dissection in irrigated farming are projected. The dynamics of landscape metrics and change processes combined in one analytical framework can facilitate understanding and detection of the relationship between ecological processes and landscape pattern. The finding of current research will provide a roadmap for improved LULC management and planning in the Talar watershed, southern coast of the Caspian Sea.

show abstract

“…The total error of corrections was estimated according to root mean standard error (RMSE), and gives 1.78 m in pixel (Giriraj et al, 2008;Yousefi et al, 2015b). Using true composite images in Landsat data (Red, Green, and Blue bands) and normalized difference water index (NDWI), the active channel for study periods by ENVI 4.8 was extracted (Haibo et al, 2011;Ko et al, 2015;Yousefi et al, 2015a). Based on 2016 Landsat OLI image, the study reach was divided into 48 meander loops.…”

Section: Data Usedmentioning

confidence: 99%

Effects of hydrological events on morphological evolution of a fluvial system

Mirzaee

Yousefi

Keesstra

et al. 2018

Journal of Hydrology

Self Cite

View full text Add to dashboard Cite

This study quantifies morphological evolution of the Dez River, Iran, from 1955 to 2016. The approach uses a sequence of Landsat images, aerial photos, and topographic maps. In addition, the hydrological data including average daily discharge and yearly maximum discharge at the Dezful hydrological station for the period (1955-2016) were used. The study reach was divided into 48 meander loops from upstream to downstream. Active channel width (w) was determined at 10 m intervals and changes assessed along the study reach of the Dez River. Morphological indices including sinuosity index; straight meander length; centerline flow length; erosion area; erodible length channel migration; centerline elongation; and radius of curvature were calculated in the reach. Results showed that the study reach of the Dez River changed dramatically in response to major floods, although the general trend is towards a narrowing of active channel width by 38% in the period 1955-2016. Results show that most of the meander loops in the study area extended and expanded. Between 1989 and 1995, all types of meander change were observed. There was a direct correlation between the frequency of hydrological events (flood days) bigger than 2-years return period and elongation of bends. 1. Introduction Rivers represent an important and dynamic landscape feature in the world's great plains and valleys. Fluvial systems in these regions are, however, under threat, coming under increasing pressure from growing populations and the demand placed on water, particularly in drier zones. River systems are used as a source of energy, food, transport and subjected to geopolitics (Surian and Rinaldi, 2003; Yousefi et al., 2017a). Fluvial ecosystems are among the most threatened ecosystems in the world, where pollution, urbanization, bio-invasion or changes in water regime results in degraded landscapes (Das et al., 2014; Luppi et al., 2009a; Thakur et al., 2012). Change in river morphology results in adjustment of other morphological and ecological characteristics until the river adapts to the new conditions (

show abstract

Accuracy assessment of land cover/land use classifiers in dry and humid areas of Iran

Cited by 36 publications

References 41 publications

Exploring Land Cover Classification Accuracy of Landsat 8 Image Using Spectral Index Layer Stacking in Hilly Region of South Korea

Exploring Land Cover Classification Accuracy of Landsat 8 Image Using Spectral Index Layer Stacking in Hilly Region of South Korea

Landscape Management through Change Processes Monitoring in Iran

Effects of hydrological events on morphological evolution of a fluvial system

Contact Info

Product

Resources

About