A comparison of the integrated fuzzy object-based deep learning approach and three machine learning techniques for land use/cover change monitoring and environmental impacts assessment

Feizizadeh, Bakhtiar; Alajujeh, Keyvan Mohammadzade; Lakes, Tobia; Blaschke, Thomas; Omarzadeh, Davoud

doi:10.1080/15481603.2021.2000350

Cited by 36 publications

(21 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recent progress in earth observation and remote sensing technologies produced satellite images with improved spatial, spectral, and temporal resolutions, which accordingly increased the demands of efferent data-driven approaches [53][54][55][56]. In this context, a number of machine learning and particularly deep learning methods were developed and proposed over the past decade [13]. Despite developing a variety of classification methods (e.g., machine learning, deep learning, etc.…”

Section: Significance Of Qadimentioning

confidence: 99%

“…It is sometimes argued that progress in technology and data analysis and increasing demand for efficient and cost-effective data-driven approaches have revolutionized Earth Observation methods [2,3]. Recent work illustrates the demand from various application fields for effective data-driven solutions [4][5][6][7][8][9][10][11][12][13]. Several efficient data-driven approaches (e.g., semi/automated and machine learning methods, deep learning, conventional network, etc.)…”

Section: Introductionmentioning

confidence: 99%

“…At the same time, novel data-driven approaches demand efficient accuracy assessment methods [13]. However, despite significant progress in image classification techniques, the development of accuracy assessment methods has not kept pace.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

QADI as a New Method and Alternative to Kappa for Accuracy Assessment of Remote Sensing-Based Image Classification

Feizizadeh

Darabi

Blaschke

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

Classification is a very common image processing task. The accuracy of the classified map is typically assessed through a comparison with real-world situations or with available reference data to estimate the reliability of the classification results. Common accuracy assessment approaches are based on an error matrix and provide a measure for the overall accuracy. A frequently used index is the Kappa index. As the Kappa index has increasingly been criticized, various alternative measures have been investigated with minimal success in practice. In this article, we introduce a novel index that overcomes the limitations. Unlike Kappa, it is not sensitive to asymmetric distributions. The quantity and allocation disagreement index (QADI) index computes the degree of disagreement between the classification results and reference maps by counting wrongly labeled pixels as A and quantifying the difference in the pixel count for each class between the classified map and reference data as Q. These values are then used to determine a quantitative QADI index value, which indicates the value of disagreement and difference between a classification result and training data. It can also be used to generate a graph that indicates the degree to which each factor contributes to the disagreement. The efficiency of Kappa and QADI were compared in six use cases. The results indicate that the QADI index generates more reliable classification accuracy assessments than the traditional Kappa can do. We also developed a toolbox in a GIS software environment.

show abstract

Section: Significance Of Qadimentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

QADI as a New Method and Alternative to Kappa for Accuracy Assessment of Remote Sensing-Based Image Classification

Feizizadeh

Darabi

Blaschke

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…NDVI is known as an efficient index for density of vegetation covers. The land use/cover (LULC) map was derived from the Landsat satellite images obtained as part of our earlier research on monitoring the LULC changes in the ULB from 1990-2020 using an integrated approach of object-based image analysis and deep learning techniques [31,38,39]. We used 150 ground control data points and their respective laboratory analysis to validate the results.…”

Section: Data Acquisitionmentioning

confidence: 99%

“…The severity of the Urmia Lake drought is intensified by other subsidiary influences of climate change, such as the rapid depletion of groundwater, decreasing groundwater quality, soil salinization, loss of soil fertility and intensive soil degradation [29,30]. The environmental issues caused by this lake drought are expected to be powerful enough to cause potential future migration and human displacement of susceptible rural communities from affected areas to other regions of the country [31,32]. Due to its increasingly critical condition, the United Nations Environment Program (UNEP) declared Urmia Lake's status as "worrying" in its Global Environmental Alert Service Bulletin in February 2012.…”

Section: Introductionmentioning

confidence: 99%

Impacts of the Urmia Lake Drought on Soil Salinity and Degradation Risk: An Integrated Geoinformatics Analysis and Monitoring Approach

et al. 2022

Self Cite

View full text Add to dashboard Cite

Recent improvements in earth observation technologies and Geographical Information System (GIS) based spatial analysis methods require us to examine the efficiency of the different data-driven methods and decision rules for soil salinity monitoring and degradation mapping. The main objective of this study was to analyze the environmental impacts of the Lake Urmia drought on soil salinity and degradation risk in the plains surrounding the hyper-saline lake. We monitored the impacts of the lake drought on soil salinity by applying spatiotemporal indices to time-series satellite images (1990–2020) in Google Earth Engine environment. We also computed the soil salinity ratio to validate the results and determine the most efficient soil salinity monitoring techniques. We then mapped the soil degradation risk based on GIS spatial decision-making methods. Our results indicated that the Urmia Lake drought is leading to the formation of extensive salt lands, which impact the fertility of the farmlands. The land affected by soil salinity has increased from 2.86% in 1990 to 16.68% in 2020. The combined spectral response index, with a performance of 0.95, was the most efficient image processing method to assess soil salinity. The soil degradation risk map showed that 38.45% of the study area has a high or very high risk of degradation, which is a significant threat to food production. This study presents an integrated geoinformation approach for time-series soil salinity monitoring and degradation risk mapping that supports future studies by comparing the efficiency of different methods as state of the art. From a practical perspective, the results also provide key information for decision-makers, authorities, and local stakeholders in their efforts to mitigate the environmental impacts of lake drought and sustain the food production to sustain the 7.3 million residents.

show abstract

Model‐coupled GRACE‐based analysis of hydrological dynamics of drying Lake Urmia and its basin

Khorrami

Ali

Sahin

et al. 2023

Hydrological Processes

View full text Add to dashboard Cite

Lake Urmia basin (LUB), in northwestern Iran, is under the influence of extreme degradation due to a number of natural and anthropogenic factors. The existence of the Lake is critical for the microclimate of the region as well as the quality of human life and wildlife, which necessitates an up-to-date and holistic analysis of its hydrological dynamics. In this premise, satellite-based terrestrial water storage (TWS) received from the Gravity Recovery and Climate Experiment (GRACE) mission was coupled with hydrometeorological modelling and assessment tools to analyse the hydrological status of the lake and its basin. As a new gap-filling approach, the Seasonal-Trend decomposition using Locally estimated scatterplot smoothing (LOESS) (STL) decomposition technique was proposed in this study to reconstruct the missing TWS data.Integrating satellite precipitation data with the Catchment Land Surface Model (CLSM) and WaterGAP model outputs, the hydrological status of the lake was investigated. The STL-based TWS turned out to concord well with the simulated TWS from the CLSM indicating the acceptable performance of the proposed technique.The findings revealed that the LUB had undergone an alarming hydrological situation from 2003 to 2021 with a total loss of 10 and 7:56 km 3 from its TWS and groundwater storage (GWS), respectively. The water level time series also indicated that the water level of the lake had diminished with an annual rate of À70 AE 21 cm=year corresponding to a total water level depletion of about 13:35 AE 3:9 m during the 2003-2021 period. The GRACE-derived TWS and GWS also agreed well with the CLSM simulations. Assessment of the extreme events of the LUB suggested that the basin suffered from a severe dry event in 2008 resulting in the depletion of its water storage and water level. It was also found that from 2003 onward, a critical hydrological setting had dominated the LUB with a negative hydrological balance of À0:96 km 3 .

show abstract

A comparison of the integrated fuzzy object-based deep learning approach and three machine learning techniques for land use/cover change monitoring and environmental impacts assessment

Cited by 36 publications

References 76 publications

QADI as a New Method and Alternative to Kappa for Accuracy Assessment of Remote Sensing-Based Image Classification

QADI as a New Method and Alternative to Kappa for Accuracy Assessment of Remote Sensing-Based Image Classification

Impacts of the Urmia Lake Drought on Soil Salinity and Degradation Risk: An Integrated Geoinformatics Analysis and Monitoring Approach

Model‐coupled GRACE‐based analysis of hydrological dynamics of drying Lake Urmia and its basin

Contact Info

Product

Resources

About