Figure 3 - uploaded by Omid Ghorbanzadeh
Content may be subject to copyright.
Source publication
The importance of Land Cover (LC) classification is recognized by an increasing number of scholars who employ LC information in various applications (i.e., address global climate change and achieve sustainable development). However, studying the roles of balancing data, image integration, and performance of different machine learning algorithms in...
Contexts in source publication
Context 1
... analysis also showed that adopting the G-SMOTE method to rebalance reference datasets substantially improved UA and PA accuracies of minority classes. As illustrated in Figure 3, RF-G-SMOTE showed the best performance in four landscapes, namely coastal, cropland, desert, and semi-arid. In comparison, the SVM-G-SMOTE obtained higher accuracies for the remaining two landscapes, including plain and mountain. ...
Context 2
... analyzing the results of SVM-RFE to choose the most critical feature for each landscape, it was revealed that NDVI, VV, and B12 bands were selected as prominent features in all six landscapes, which could confirm their importance in LC classification [16]. In contrast, NDBI, which consider as an essential index for extracting built-up areas [23], was only introduced as a critical feature in three landscapes (namely plain, semi-arid, Figure 3. Impact of G-SMOTE on the overall accuracy of the generated LC maps. ...
Similar publications
Microclimate ecology is attracting renewed attention because of its fundamental importance in understanding how organisms respond to climate change. Many hot issues can be investigated in desert ecosystems, including the relationship between species distribution and environmental gradients (e.g., elevation, slope, topographic convergence index, and...
Conservation efforts have traditionally focused on biodiversity hotspots, overlooking the essential ecological roles and ecosystem services provided by cold spots, the regions that harbour relatively low species diversity. In this study, we used a novel plant species database aggregated at 1˚ grid resolution to predict present and future plant spec...
In essence, targeting mineralization necessitates exact structural delineation and thorough lithological mapping. The latter is still a challenge for geologists and its lack hinders meticulous exploration for various mineralizations. Here we show for the first time over a case study from Arabian Nubian Shield (ANS), the application of hyperspectral...
The study examined three machine learning algorithms (MLAs): random forest (RF), support vector machine (SVM), and artificial neural networks (ANN) for generating flood susceptibility maps in two watersheds in Jordan. Both were selected because they represent two climatic regimes: desert and mountainous areas. Because of a shortage of past floods l...
Desert vegetation is undergoing complex and diverse changes due to global climate change and human activities. To effectively utilize water resources and promote ecological recovery in desert areas, it is necessary to clarify the main driving mechanisms of vegetation growth in these regions. In this study, based on MODIS and Landsat 8 remote sensin...
Citations
... In LULC classification, various SMOTE variants have been implemented to address imbalanced data. For instance, G_SMOTE [14] has been applied by Douzas et al. [15] and Ebrahimy et al. [16], while kmeans_SMOTE [17] has been explored by Fonseca et al. [12]. Standard SMOTE and its variant, ADASYN, have also been utilized [18]. ...
This study addresses the persistent challenge of class imbalance in land use and land cover (LULC) classification within the Shihmen Reservoir watershed in Taiwan, where LULC is used to map the Cover Management factor (C-factor). The dominance of forests in the LULC categories leads to an imbalanced dataset, resulting in poor prediction performance for minority classes when using machine learning techniques. To overcome this limitation, we applied the Synthetic Minority Over-sampling Technique (SMOTE) and the 90-model SMOTE-variants package in Python to balance the dataset. Due to the multi-class nature of the data and memory constraints, 42 models were successfully used to create a balanced dataset, which was then integrated with a Random Forest algorithm for C-factor classification. The results show a marked improvement in model accuracy across most SMOTE variants, with the Selected Synthetic Minority Over-sampling Technique (Selected_SMOTE) emerging as the best-performing method, achieving an overall accuracy of 0.9524 and a sensitivity of 0.6892. Importantly, the previously observed issue of poor minority class prediction was resolved using the balanced dataset. This study provides a robust solution to the class imbalance issue in C-factor classification, demonstrating the effectiveness of SMOTE variants and the Random Forest algorithm in improving model performance and addressing imbalanced class distributions. The success of Selected_SMOTE underscores the potential of balanced datasets in enhancing machine learning outcomes, particularly in datasets dominated by a majority class. Additionally, by addressing imbalance in LULC classification, this research contributes to Sustainable Development Goal 15, which focuses on the protection, restoration, and sustainable use of terrestrial ecosystems.
... We used 1000 reference samples (700 training and 300 validation samples). Finally, this study used a random forest classifier, known for its high performance in various land cover mapping tasks (Ebrahimy et al. 2021;Naboureh et al. 2021), to classify Sichuan into three classes at a 30-m resolution. The generated map showed reasonable accuracy for both the overall and individual classes. ...
Air pollution is the greatest health risk to human health, as acknowledged in several Sustainable Development Goals (SDG), such as SDG-1, SDG-3, SDG-7, and SDG-11. Despite global comprehension of the positive effect of Green Space Coverage (GSC) on mitigating air pollution, investigating the impact of different GSC types has received little attention. Here, we utilized multiple air pollution data and a cloud-computing platform to examine the role of different GSC types in mitigating NO2 and PM2.5 pollutants across 2019 and 2022 in Sichuan Province, China. We classified GSC areas into tall GSC and short GSC classes, taking into account the recognized importance of vegetation height in prior studies. Our analysis revealed that tall GSCs exhibit lower pollutant levels across all areas studied, indicating a potential correlation between GSC height and pollution mitigation efficacy. Furthermore, in high human activity areas, while tall GSC emerged as effective sinks for PM 2.5 compared to short GSC (25% and 20% lower average annual in 2019 and 2022, respectively), their performance in reducing NO2 pollutant levels was relatively limited (9% and 4% lower average annual in 2019 and 2022, respectively). These findings can contribute to urban planning and environmental management.
... The European Space Agency's (ESA) launch of the Sentinel-2 mission has greatly improved the capabilities for LULC classification with high-resolution multitemporal optical imagery. High spatial resolution of up to 10 meters, frequent revisit times, and multiple spectral bands make Sentinel-2 ideal for capturing dynamic changes in land surfaces, establishing it as a cornerstone in the remote sensing community (Nguyen et al., 2020;Ebrahimy et al., 2021;Bhatti & Tripathi, 2021;Ghorbanzadeh et al., 2023). ...
Multitemporal imagery offers a critical advantage by capturing seasonal variations, which are essential for differentiating between land use and land cover (LULC) types. While these types may appear similar when examined at one specific time, they exhibit distinct phenological patterns across different seasons. This temporal depth is crucial for enhancing model accuracy, particularly in heterogeneous landscapes where LULC transitions are frequent and complex. This paper made use of spectral bands and indices of Sentinel-2 from April to September 2020 were utilized for LULC classification using two advanced machine learning models: Random forest (RF) and support vector machine (SVM). The spectral indices include the normalized difference vegetation index (NDVI), normalized difference built-up index (NDBI), and modified normalized water index (MNDWI). The dataset was divided into four temporal feature sets: April-May, June-July, August-September, and the entire period from April-September. For each two-month period, the median values of the spectral bands and indices were used. Both models were evaluated based on overall accuracy, F1-score, Kappa coefficient, precision, and recall. Results indicate that incorporating temporal features enhanced the performance of the chosen models, with overall accuracy increasing from 82.4% to 94.03% for RF and from 75.4% to 93.54% for SVM. Additionally, the RF algorithm demonstrated higher accuracy than the SVM model across various time periods, with notable increases in F1 scores, Kappa statistic, precision, and recall. These improvements underscore the ability of the models to leverage rich temporal and spectral data provided by Sentinel-2 for accurate LULC classification. This study highlights the importance of incorporating temporal dynamics in remote sensing applications to enhance the precision and reliability of LULC classification.
... (6) Last, using 30 m Shuttle Radar Topography Mission (SRTM) data, geometric errors and distortions in images were corrected. Following the classification system proposed by Ebrahimy et al. [48], since S2 images include both 10 and 20 m spectral bands, all 20 m resolution bands (5, 6, 7, 8A, 11, and 12) were resampled to 10 m using the nearest neighbor method. A co-registration method was finally conducted to adjust pixel values of S1 and S2 images. ...
Reliable and up-to-date training reference samples are imperative for land cover (LC) classification. However, such training datasets are not always available in practice. The sample migration method has shown remarkable success in addressing this challenge in recent years. This work investigated the application of Sentinel-1 (S1) and Sentinel-2 (S2) data in training sample migration. In addition, the impact of various spectral bands and polarizations on the accuracy of the migrated training samples was also assessed. Subsequently, combined S1 and S2 images were classified using the Support Vector Machines (SVM) and Random Forest (RF) classifiers to produce annual LC maps from 2017 to 2021. The results showed a higher accuracy (98.25%) in training sample migrations using both images in comparison to using S1 (87.68%) and S2 (96.82%) data independently. Among the LC classes, the highest accuracy in migrated training samples was found for water, built-up, bare land, grassland, cropland, and wetland. Inquiries on the efficiency of different spectral bands and polarization used in training sample migration showed that bands 4 and 8 and VV polarization in the water class were more important, while for the wetland class, bands 5, 6, 7, 8, and 8A together with VV polarization showed superior performance. The results showed that the RF classifier provided better performance than the SVM (higher overall, producer, and user accuracy). Overall, our findings suggested that shared use of S1 and S2 data can be used as a suitable means for producing up-to-date and high-quality training samples.
... RF has become a popular choice for large-scale LC classification due its ability to handle complex and high-dimensional data 31 . An investigation, which focused on six different landscapes in the Corridor, found that RF had the best performance among several well-known supervised classification algorithms 32 . In addition, our research 19 revealed that using an adaptive RF-based classification method can effectively address the impact of climate variations on LC classification accuracy. ...
Land Cover (LC) maps offer vital knowledge for various studies, ranging from sustainable development to climate change. The China Central-Asia West-Asia Economic Corridor region, as a core component of the Belt and Road initiative program, has been experiencing some of the most severe LC change tragedies, such as the Aral Sea crisis and Lake Urmia shrinkage, in recent decades. Therefore, there is a high demand for producing a fine-resolution, spatially-explicit, and long-term LC dataset for this region. However, except China, such dataset for the rest of the region (Kyrgyzstan, Turkmenistan, Kazakhstan, Uzbekistan, Tajikistan, Turkey, and Iran) is currently lacking. Here, we constructed a historical set of six 30-m resolution LC maps between 1993 and 2018 at 5-year time intervals for the seven countries where nearly 200,000 Landsat scenes were classified into nine LC types within Google Earth Engine cloud computing platform. The generated LC maps displayed high accuracies. This publicly available dataset has the potential to be broadly applied in environmental policy and management.
... The provinces of Gilan, Mazandarn, Kohgiluyeh and Boyer-Ahmad, Golestan, Alborz, Ardabil, and North Khorasan have the smallest amounts of agricultural land experiencing critical drought conditions during the study period. Researchers have explored the application of land cover classification to drought conditions (i.e., global climate change and sustainable development) (Eskandari et al. 2020;Phiri et al. 2020;Ebrahimy et al. 2021). ...
This study spatially monitored drought in Iran using drought indicators. Four drought indicators measured from 2016 to 2020 were used: temperature condition index (TCI), vegetation condition index (VCI), vegetation health index (VHI), and precipitation condition index (PCI). Moreover, a standardized precipitation index (SPI) was prepared using rainfall measurements from 1989 to 2019. The TCI revealed that most of Iran was classified as having “severe drought” in 2020. The highest value of VCI showed for northern Iran, which belongs to the class without drought. The VHI indicated that vegetation stress increased over the study period throughout the region, and areas of severe and moderate drought reached their greatest extents in the aforementioned years. Significant droughts occurred in central, eastern, and southeastern Iran, and mild droughts occurred in northern Iran. The PCI indicated that rainfall amounts have diminished in most of the country over the period of study. The 30-year SPI showed that northern Iran received fine rain and the region has parts that can be classified as either extremely wet or very wet. However, most of the country was extremely dry and severely dry. The analysis of the VHI index for agricultural plants showed that 27.71% of Iran's agricultural regions, including the provinces of Razavi Khorasan, Hamadan, and Khozestan, experienced “critical drought” conditions. The study provides guidance for the selection of the most useful drought-monitoring indicators and can enable a more thorough understanding of drought in arid and semiarid regions.
... To compare the efficiency of Fuzzy-OBIA-DL with machine learning algorithms, we also tested the following three machine learning (ML) algorithms: support vector machine (SVM), random forest (RF), and classification and regression tree (CART). Due to the popularity of the ML algorithms for satellite image classification, several ML algorithms have been proposed and applied in earlier research (Talukdar et al. 2020;Ebrahimy et al. 2021;Ghorbanzadeh et al. 2020;Omarzadeh et al. 2021;Valizadeh Kamran et al. 2021). The ML algorithms allow analyzing complex data spaces (e.g. ...
Recent improvements in the spatial, temporal, and spectral resolution of satellite images necessitate (semi-)automated classification and information extraction approaches. Therefore, we developed an integrated fuzzy object-based image analysis and deep learning (FOBIA-DL) approach for monitoring the land use/cover (LULC) and respective changes and compared it to three machine learning (ML) algorithms, namely the support vector machine (SVM), random forest (RF), and classification and regression tree (CART). We investigated LULC impacts on drought by analyzing Landsat satellite images from 1990 to 2020 for the Urmia Lake area in northern Iran. In the FOBIA-DL approach, following the initial segmentation steps, object features were identified for each LULC class. We then derived their respective attributes using fuzzy membership functions and deep convolutional neural networks (DCNNs), a deep learning method. The Fuzzy Synthetic Evaluation and Dempster-Shafer Theory (FSE-DST) was also applied to validate and carry out the spatial uncertainties. Our results indicate that the FOBIA-DL, with an accuracy of 90.1% to 96.4% and a spatial certainty of 0.93 to 0.97, outperformed the other approaches, closely followed by the SVM. Our results also showed that the integration of Fuzzy-OBIA and DCNNs could improve the strength and robustness of the OBIA’s decision rules, while the FSE-DST approach notably improved the spatial accuracy of the object-based classification maps. While object-based image analysis (OBIA) is already considered a paradigm shift in GIScience, the integration of OBIA with fuzzy and deep learning creates more flexibility and robust OBIA decision rules for image analysis and classification. This research integrated popular data-driven approaches and developed a novel methodology for image classification and spatial accuracy assessment. From the environmental perspective, the results of this research support lake restoration initiatives by decision-makers and authorities in applications such as drought mitigation, land use management, and precision agriculture programs.
