Supervised Classification Techniques

Richards, John A.

doi:10.1007/978-3-642-30062-2_8

Cited by 44 publications

(26 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data were analyzed by using a combination of techniques including Remote Sensing, Geographic Information System-based processing, logistic regressions, and descriptive statistics. The LULC classification was done by employing a supervised maximum likelihood algorithm classification approach which is generally recognized as the popular classifier technique (Booth & Oldfield 1989;Hogland et al 2013;Richards 2013). The supervised classification approach requires prior knowledge of the target area to set training sites and use of spectral information contained in individual pixels to generate LULC classes.…”

Section: Discussionmentioning

confidence: 99%

Drivers of Land Use-Land Cover Changes in the Central Rift Valley of Ethiopia

Bekele¹,

Wu²,

Yirsaw³

2019

JSM

View full text Add to dashboard Cite

Land use-land cover change (LULCC) is driven by the interplay of forcing factors that act at global, regional, and local levels. Previous studies investigated mainly the basic socioeconomic drivers of LULCC. However, these studies less considered climate change vulnerability as a potential driver. Hence, this study is aimed to assess LULCC drivers in more fragile and dynamic landscapes of the East African Rift Valley region for the period of 1986-2016. We used a combination of Remote Sensing, Geographic Information System, logistic regression, and descriptive statistics to quantify and analyze the data. Image analysis results indicated that during the overall study period (1986-2016), grass/ grazing land, agricultural land, and bare land have increased by 124%, 42%, and 34% respectively, whereas scattered acacia woodland, bush/shrubland, and swampy/marshy land have declined by 52%, 50%, and 31%, in that order. This image-derived change trend is in line with farmers' perceived results. The top most influential drivers of LULCC includes population growth (95%), fuelwood extraction (93%), agricultural land expansion (92%), charcoal making (92%), climate change/recurrent drought (79%), and overgrazing (71%) in descending order of percentage of respondents. Education level and age of farmers significantly (p<0.05) affected their perception towards less perceived drivers. Hence, in order to reduce the adverse socio-environmental impacts of spectacular LULCC in the region, policy and decision makers need to take into account such principal drivers, particularly population growth and climate change.

show abstract

Section: Discussionmentioning

confidence: 99%

Drivers of Land Use-Land Cover Changes in the Central Rift Valley of Ethiopia

Bekele¹,

Wu²,

Yirsaw³

2019

JSM

View full text Add to dashboard Cite

show abstract

“…Supervised classification: Supervised classification is one of the techniques most frequently used for quantitative analysis of RS data [44]. Maximum Likelihood Classification (MLC) is a well-known and most widely used parametric classification algorithm [45][46][47].…”

Section: Image Processing and Extraction Of Lulcmentioning

confidence: 99%

Land Use and Land Cover Dynamics and Assessing the Ecosystem Service Values in the Trans-Boundary Gandaki River Basin, Central Himalayas

et al. 2018

View full text Add to dashboard Cite

Land use and land cover is a fundamental variable that affects many parts of social and physical environmental aspects. Land use and land cover changes (LUCC) has been known as one of the key drivers of affecting in ecosystem services. The trans-boundary Gandaki River Basin (GRB) is the part of Central Himalayas, a tributary of Ganges mega-river basin plays a crucial role on LUCC and ecosystem services. Due to the large topographic variances, the basin has existed various land cover types including cropland, forest cover, built-up area, river/lake, wetland, snow/glacier, grassland, barren land and bush/shrub. This study used Landsat 5-TM (1990), Landsat 8-OLI (2015) satellite image and existing national land cover database of Nepal of the year 1990 to analyze LUCC and impact on ecosystem service values between 1990 and 2015. Supervised classification with maximum likelihood algorithm was applied to obtain the various land cover types. To estimate the ecosystem services values, this study used coefficients values of ecosystem services delivered by each land cover class. The combined use of GIS and remote sensing analysis has revealed that grassland and snow cover decreased from 10.62% to 7.62% and 9.55% to 7.27%, respectively compared to other land cover types during the 25 years study period. Conversely, cropland, forest and built-up area have increased from 31.78% to 32.67%, 32.47–33.22% and 0.19–0.59%, respectively in the same period. The total ecosystem service values (ESV) was increased from 50.16 × 108 USD y−1 to 51.84 × 108 USD y−1 during the 25 years in the GRB. In terms of ESV of each of land cover types, the ESV of cropland, forest, water bodies, barren land were increased, whereas, the ESV of snow/glacier and grassland were decreased. The total ESV of grassland and snow/glacier cover were decreased from 3.12 × 108 USD y−1 to 1.93 × 108 USD y−1 and 0.26 × 108 USD y−1 to 0.19 × 108 USD y−1, respectively between 1990 and 2015. The findings of the study could be a scientific reference for the watershed management and policy formulation to the trans-boundary watershed.

show abstract

“…The Senetal satellite datasets were used and various land use land cover categories were delineated using the supervised classification methods by providing signatures of different land use categories (Ahmad et al 2018b). Supervised classification is an approach widely used for the quantitative analysis of remote sensing image data using the concept of segmenting the spectral domain into regions (Richards, 2013). The various land use land cover categories such as trees, tree mixed, agriculture, settlement, water and open area were delineated which is given in figure 7.…”

Section: Land Use/ Land Cover (Lulc) Mapmentioning

confidence: 99%

Assessment of remote sensing and GIS application in identification of land suitability for agroforestry: A case study of Samastipur, Bihar, India

Ahmad¹,

Uddin²,

Goparaju³

2018

Contemporary Trends in Geoscience

View full text Add to dashboard Cite

Agroforestry provides the foundation for climate-smart agriculture to withstand the extreme weather events. The aim of the present study was to identify the land of Samastipur, Bihar, India for agroforestry, based on GIS modeling concept utilizing various ancillary (soil fertility) and satellite data (DEM, wetness, NDVI and LULC) sets. This was achieved by integrating various thematic layers logically in GIS domain. Agroforestry suitability maps were generated for the Samastipur district of Bihar, India which showed 48.22 % as very high suitable, 22.83 % as high suitable, 23.32% as moderate suitable and 5.63% as low suitable. The cross evaluation of agroforestry suitability with LULC categories revealed that the 86.4 % (agriculture) and 30.2% (open area) of land fall into a very high agroforestry suitability category which provides the huge opportunity to harness agroforestry practices if utilized scientifically. Such analysis/results will certainly assist agroforestry policymakers and planner in the state of Bihar, India to implement and extend it to new areas. The potentiality of Remote Sensing and GIS can be exploited in accessing suitable land for agroforestry which will significantly help to rural poor people/farmers in ensuring food and ecological security, resilience in livelihoods.

show abstract

Supervised Classification Techniques

Cited by 44 publications

References 41 publications

Drivers of Land Use-Land Cover Changes in the Central Rift Valley of Ethiopia

Drivers of Land Use-Land Cover Changes in the Central Rift Valley of Ethiopia

Land Use and Land Cover Dynamics and Assessing the Ecosystem Service Values in the Trans-Boundary Gandaki River Basin, Central Himalayas

Assessment of remote sensing and GIS application in identification of land suitability for agroforestry: A case study of Samastipur, Bihar, India

Contact Info

Product

Resources

About