Land Cover Classification Accuracy Research Articles

Application of Sentinel-2A Images for Land Cover Classification Using NDVI in Jember Regency

The advancement of remote sensing technology has led to the development of sophisticated image processing methods that yield highly accurate land cover classification information, minimizing misinterpretations. The Normalized Difference Vegetation Index (NDVI) is a widely utilized method in remote sensing for measuring green vegetation. A significant portion of the Jember Regency area is covered by vegetation. This study aimed to identify various land cover types in the Jember Regency area, quantify the area for each classification, and establish the NDVI value ranges for each type of cover. Sentinel-2 was employed as the primary data source, and the NDVI method was utilized for land cover classification in the Jember Regency. The region exhibited diverse land cover types. Data from Sentinel-2A captured in June and October 2019 were chosen due to their accessibility, open-source nature, and adequate spectral, spatial, and temporal resolution. The classification in this study encompassed five classes: water bodies, settlements, dry fields, irrigated paddy fields, and forests. Error analysis was conducted using a confusion matrix with the Overall and Kappa algorithms. The accuracy results for June indicated a Kappa Accuracy of 37.7% and Overall Accuracy of 54.5%. In October, the Kappa Accuracy increased to 39.9%, and the Overall Accuracy reached 56.5%. In conclusion, the NDVI method did not meet the criteria for accurately interpreting land cover classification.

Improving land cover classification accuracy of Sentinel-2 images: a systematic review of articles between 2015 and 2021

Improving land cover classification accuracy of Sentinel-2 images: a systematic review of articles between 2015 and 2021

Multi-Scale Feature Fusion Network with Symmetric Attention for Land Cover Classification Using SAR and Optical Images

The complementary characteristics of SAR and optical images are beneficial in improving the accuracy of land cover classification. Deep learning-based models have achieved some notable results. However, how to effectively extract and fuse the unique features of multi-modal images for pixel-level classification remains challenging. In this article, a two-branch supervised semantic segmentation framework without any pretrained backbone is proposed. Specifically, a novel symmetric attention module is designed with improved strip pooling. The multiple long receptive fields can better perceive irregular objects and obtain more anisotropic contextual information. Meanwhile, to solve the semantic absence and inconsistency of different modalities, we construct a multi-scale fusion module, which is composed of atrous spatial pyramid pooling, varisized convolutions and skip connections. A joint loss function is introduced to constrain the backpropagation and reduce the impact of class imbalance. Validation experiments were implemented on the DFC2020 and WHU-OPT-SAR datasets. The proposed model achieved the best quantitative values on the metrics of OA, Kappa and mIoU, and its class accuracy was also excellent. It is worth mentioning that the number of parameters and the computational complexity of the method are relatively low. The adaptability of the model was verified on RGB–thermal segmentation task.

Fine-resolution land cover mapping over large and mountainous areas for Lāna‘i, Hawaii using posterior probabilities, and expert knowledge

ABSTRACT The task of accurately mapping species-specific vegetation cover in remote and topographically complex regions like those found in Hawaiʻi presents unique challenges. This study leverages a machine learning approach to accurately classify vegetation into fine species-specific classes across the island of Lāna‘i, Hawaii, offering a novel methodology for tackling such challenges. Utilizing high-resolution WordView-2 satellite imagery, a neural network classifier and a custom lidar-based geometric correction, we introduced two new approaches to refine our high-resolution land cover classifications. This included the implementation of prior-based adjustments to class posterior probabilities to enhance land cover classification accuracy. Moreover, we developed mixed hierarchical classification maps that use class posterior probabilities to identify, at the pixel level, the finest land cover class that meets a user-defined confidence threshold. The resulting high-resolution land cover map for Lāna‘i captures the rich diversity and distribution of native and invasive plant species with high overall accuracy, generally exceeding 95%, based on independent ground control data. The capacity to produce wall-to-wall species-level vegetation maps provides a new window into monitoring vegetation dynamics on Lāna‘i and similarly remote and topographically complex regions, and contributes to our broader understanding of ecosystem responses to invasive species, climatic changes, and land management practices such as erosion and sediment control planning. Our approach offers a blueprint for similar efforts in other complex and remote ecosystems.

Spatio-temporal analysis of urban expansion using remote sensing data and GIS for the sustainable management of urban land: the case of Burayu, Ethiopia

PurposeThe purpose of this study is to analyze the horizontal expansion of Burayu Town between 1990 and 2020. The study typically acts as a baseline for integrated spatial planning in small- and medium-sized towns, which will help to plan sustainable utilization of land.Design/methodology/approachLandsat5-TM, Landsat7 ETM+, Landsat5 TM and Landsat8 OLI were used in the study, along with other auxiliary data. The LULC map classifications were generated using the Random Forest Package from the Comprehensive R Archive Network. Post-classification, spatial metrics, and per capita land consumption rate were used to understand the manner and rate of expansion of Burayu Town. Focus group discussions and key informant interviews were also used to validate land use classes through triangulation.FindingsThe study found that the built-up area was the most dynamic LULC category (85.1%) as it increased by over 4,000 ha between 1990 and 2020. Furthermore, population increase did not result in density increase as per capita land consumption increased from 0.024 to 0.040 during the same period.Research limitations/implicationsAs a result of financial limitations, there were no high-resolution satellite images available, making it challenging to pinpoint the truth as it is on the ground. Including senior citizens in the study region allowed this study to overcome these restrictions and detect every type of land use and cover.Practical implicationsData on urban growth are useful for planning land uses, estimating growth rates and advising the government on how best to use land. This can be achieved by monitoring and reviewing development plans using satellite imaging data and GIS tools.Originality/valueThe use of Random Forest for image classification and the employment of local knowledge to validate the accuracy of land cover classification is a novel approach to properly customize remote sensing applications.

Hyperspectral Tensor Completion Using Low-Rank Modeling and Convex Functional Analysis.

Hyperspectral tensor completion (HTC) for remote sensing, critical for advancing space exploration and other satellite imaging technologies, has drawn considerable attention from recent machine learning community. Hyperspectral image (HSI) contains a wide range of narrowly spaced spectral bands hence forming unique electrical magnetic signatures for distinct materials, and thus plays an irreplaceable role in remote material identification. Nevertheless, remotely acquired HSIs are of low data purity and quite often incompletely observed or corrupted during transmission. Therefore, completing the 3-D hyperspectral tensor, involving two spatial dimensions and one spectral dimension, is a crucial signal processing task for facilitating the subsequent applications. Benchmark HTC methods rely on either supervised learning or nonconvex optimization. As reported in recent machine learning literature, John ellipsoid (JE) in functional analysis is a fundamental topology for effective hyperspectral analysis. We therefore attempt to adopt this key topology in this work, but this induces a dilemma that the computation of JE requires the complete information of the entire HSI tensor that is, however, unavailable under the HTC problem setting. We resolve the dilemma, decouple HTC into convex subproblems ensuring computational efficiency, and show state-of-the-art HTC performances of our algorithm. We also demonstrate that our method has improved the subsequent land cover classification accuracy on the recovered hyperspectral tensor.

Impact of land use/land cover changes on evapotranspiration and model accuracy using Google Earth engine and classification and regression tree modeling

This research uses a Classification and Regression Tree (CART) model with Google Earth Engine (GEE) to assess the winter season’s land cover and change detection mapping impact on the evapotranspiration (crop water requirement) parameters. Winter seasons, crucial for agricultural planning, and irrigation water requirement challenges in accurately mapping land cover and detecting changes due to the dynamic nature of farming practices during this period. In this study, Landsat-8 OLI images have been combined to map Land use and Land cover (LULC) and other change detection mapping in Akola Block, Maharashtra, India, during the 2018–2022 winter season. As an discoverer researcher that found detailed information of LULC classes during last 2018 to 2022 winter seasons, the use of the CART model in combination with a cloud-computing GEE demonstrates to be a practical approach for accurate land cover classification and change detection maps to create a pixel-based winter seasons information of study area. The novelty of this study lies in its innovative use of GEE, a powerful platform for remote sensing and geospatial analysis, to create LULC maps with remarkable accuracy. Achieving a 100% training accuracy across the four years under consideration is an exceptional feat, highlighting the reliability and stability of the methodology. Furthermore, the validation accuracy values, ranging from 89 to 94% for the winter seasons of 2018 to 2022, underscore the robustness of this approach. Such consistently high accuracy in mapping LULC over time is a groundbreaking achievement and offers a new dimension to the field of hydrology. For the hydrological community, the implications of this study are profound. Accurate LULC mapping and change detection provide critical data for modeling and analyzing the effects of land use changes on water resources, watershed management, and water quality. The User, Kappa, and Producer accuracy metrics used in this research highlight the model’s performance and its suitability for hydrological applications. These accurate LULC maps can aid in the development of hydrological models, forecasting, and decision-making processes, ultimately contributing to more effective water resource management and environmental conservation. In summary, this study’s innovative use of GEE, its remarkable accuracy in LULC mapping, and its relevance to the hydrological community demonstrate the potential for advanced remote sensing and geospatial tools to significantly improve our understanding of land use changes and their implications for water resources and environmental management.

SEMANTIC KNOWLEDGE EMBEDDING DEEP LEARNING NETWORK FOR LAND COVER CLASSIFICATION

Abstract. Land cover classification is essential basic information and key parameters for environmental change research, geographical and national monitoring, and sustainable development planning. Deep learning can automatically and multi-level extract the features of complex features, which has been proven to be an effective method for information extraction. However, one of the major challenges of deep learning is its poor interpret-ability, which makes it difficult to understand and explain the reasoning behind its classification results. This paper proposes a deep cross-modal coupling model (CMCM) for integrating semantic features and visual features. The representation of knowledge map is indicatively introduced into remote sensing image classification. Compared to previous studies, the proposed method provides accurate descriptions of the complex semantic objects within a complex land cover environment. The results showed that the integration of semantic knowledge improved the accuracy and interpret-ability of land cover classification.

A SHALLOW NEURAL NETWORK MODEL FOR URBAN LAND COVER CLASSIFICATION USING VHR SATELLITE IMAGE FEATURES

Abstract. Recently, image classification techniques using neural networks have received considerable attention in sustainable urban development, since their applications have an extreme effect on building distribution, infrastructural networks, and water resource management. In this research, a back-propagation shallow neural network model is presented for very high resolution satellite image classification in urban environments. Workflow procedures consider selecting and collecting data, preparing required study areas, extracting distinctive features, and applying the classification process. Visual interpretation is performed to identify observed land cover classes and detect distinctive features in the urban environment. Pre-processing techniques are implemented to present the used images in a more suited form for the classification techniques. A shallow neural network model (supported by MathWorks MATLAB environment) is successfully applied and results are evaluated. The proposed model is tested for classifying both WorldView-2 and WorldView-3 multispectral images with different spatial and spectral characteristics to check the model’s applicability to various kinds of satellite imagery and different study areas. Model outcomes are compared to two well-known classification methods; the Nearest Neighbour object-based method and the Maximum Likelihood pixel-based classifier, to validate and check the model stability. The overall accuracy achieved by the proposed model is 86.25% and 83.25%, while the nearest neighbour approach has obtained 84.50% and 82.75%, and the maximum likelihood classifier has accomplished 82.50% and 80.25% for study area 1 and study area 2 respectively. Obtained results indicate that the developed shallow neural network model achieves a promising accuracy for urban land cover classification in comparison with the standard techniques.

Assessing cropland disagreement in Tanzania using machine learning methods with Sentinel-2 and Planet Scope imagery

ABSTRACT East Africa faces major land use pressures arising from the harsh effects of climate change and fast-increasing human populations. Recent efforts to increase food production in this region have focused on cropland mapping and approaches to improve yield. Generating accurate cropland maps provides critical support to inform policy, investment, and logistical decisions that address food security, but cropland disagreement assessment studies are lacking mostly due to the unavailability of fine-scale data. Understanding the inconsistent nature of pre-existing global cropland mapping products necessitates disagreement assessment as the most valuable urgent tool to align agricultural resource allocation and policy realignment to tackle food security challenges. This study evaluates cropland disagreement in Kilombero and Ulanga districts in Morogoro, Tanzania based on land use land cover (LULC) maps generated using machine learning methods and high-resolution PlanetScope and Sentinel-2 data sets acquired from 1 January 2017 to 31 December 2018. We created time-variant median composites and performed a three-step assessment; (i) generating multi-class LULC maps in Google Earth Engine, (ii) generating croplands and non-cropland LULC classes from the multi-class images, and (iii) assessing cropland disagreement based on GIS conditional statements. Results show minimal disagreement in mapping cropland based on higher-resolution PlanetScope and Sentinel-2. Results based on the Random Forests and Support Vector Machines relatively outperform those from Classification and Regression Trees achieving 93%, 89%, and 83% with PlanetScope and 91%, 86%, and 76% with Sentinel-2, respectively. On average, 73% of the pixels were consistently classified while 27% were misclassified between cropland and non-cropland classes. Prior unsupervised land cover cluster analysis may substantiate the quality of sampling for accurate land cover classification. Our results indicate that in data-sparse regions where confirmatory ground truth data are lacking, remotely sensed cropland identification may be more successful using Support Vector Machines on newer high-resolution imagery, thereby improving crop forecasts.

Exploring Bangladesh's Soil Moisture Dynamics via Multispectral Remote Sensing Satellite Image

Accurate monitoring and mapping of soil moisture are essential for sustainable agricultural practices, water resource management, and climate studies. This study aims to explore the mapping of soil moisture in Bangladesh using multispectral remote-sensing satellite images. The purpose of this study is to prepare a map of soil moisture aiming to help government authorities in developing agricultural activities to accelerate the sustainable development of the rural economy in Bangladesh. A total of 14 Landsat scenes of paths 135-139 and rows 42-46 covers the entire Bangladesh. Thus, a set of Landsat imagery (a total of 14 scenes) for the year 2022 was used in this study to map the soil moisture of Bangladesh through the application of Geographical Information System (GIS) and Remote Sensing. Satellite Image preprocessing, correction, and analysis were done with ENVI software (version 5.1, developed by Research Systems, Inc., USA) and the ArcGIS software (version 10.6, developed by Environmental Systems Research Institute, USA). For the study of the long-term variation of soil moisture over Bangladesh and its seasonal characteristics, a soil moisture map can be used. In addition, to improve the climate model over Bangladesh, an up-to-date soil moisture map will be very helpful. The objective of this study is to provide accurate and detailed up-to-date spatial soil moisture information at reduced cost and time which is essential for environment modeling, risk assessment, decision-making for different government agencies and development partners, and help toward socio-economic development. In this study, the map shows soil moisture as very wet, wet, dry, and very dry soils of Bangladesh. The overall land cover classification accuracy was 92.56%, with a Kappa value 0.90 for Random Forest and the overall soil classification accuracy was 87.27%, with a Kappa value 0.858 for maximum likelihood classification indicating good consistency.

Assessing the Influence of Land Cover and Climate Change Impacts on Runoff Patterns Using CA-ANN Model and CMIP6 Data

Dhaka city is experiencing rapid land cover changes, and the effects of climate change are highly visible. Investigating their combined influence on runoff patterns is vital for sustainable urban planning and water resources management. In this work, multi-date land cover classification was performed using a random forest (RF) algorithm. To validate accuracy of land cover classification, an assessment was conducted by employing kappa coefficient, which ranged from 85 to 96%, indicating a high agreement between classified images and the reference dataset. Future land cover changes were forecasted with cellular automata-artificial neural network (CA-ANN) model. Further, soil conservation service -curve number (SCS-CN) rainfall-runoff model combined with CMIP6 climate data was employed to assess how changes in land cover impact runoff within Dhaka metropolitan development plan (DMDP) area. Over the study period (2020–2100), substantial transformations of land cover were observed, i.e., built-up areas expanded to 1146.47 km2 at the end of 2100, while agricultural areas and bare land diminished considerably. Consequently, monsoon runoff increased from 350.14 to 368.24 mm, indicating elevated hydrological responses. These findings emphasized an intricate interplay between urban dynamics and climatic shifts in shaping runoff patterns, underscoring urgency of incorporating these factors into urban planning strategies for sustainable water resources management in a rapidly growing city such as Dhaka.

Deep Spatial Graph Convolution Network with Adaptive Spectral Aggregated Residuals for Multispectral Point Cloud Classification

Over an extended period, considerable research has focused on elaborated mapping in navigation systems. Multispectral point clouds containing both spatial and spectral information play a crucial role in remote sensing by enabling more accurate land cover classification and the creation of more accurate maps. However, existing graph-based methods often overlook the individual characteristics and information patterns in these graphs, leading to a convoluted pattern of information aggregation and a failure to fully exploit the spatial–spectral information to classify multispectral point clouds. To address these limitations, this paper proposes a deep spatial graph convolution network with adaptive spectral aggregated residuals (DSGCN-ASR). Specifically, the proposed DSGCN-ASR employs spatial graphs for deep convolution, using spectral graph aggregated information as residuals. This method effectively overcomes the limitations of shallow networks in capturing the nonlinear characteristics of multispectral point clouds. Furthermore, the incorporation of adaptive residual weights enhances the use of spatial–spectral information, resulting in improved overall model performance. Experimental validation was conducted on two datasets containing real scenes, comparing the proposed DSGCN-ASR with several state-of-the-art graph-based methods. The results demonstrate that DSGCN-ASR better uses the spatial–spectral information and produces superior classification results. This study provides new insights and ideas for the joint use of spatial and spectral information in the context of multispectral point clouds.

ASSESSMENT OF IMAGE CLASSIFICATION ALGORITHMS FOR LAND COVER CLASSIFICATIONS IN TULLY, NY

Abstract. The identification, delineation, and mapping of landcover is integral for resource management and planning as it establishes a baseline for thematic mapping and change detection analysis. The availability of high-resolution satellite imagery and the development of machine learning algorithms have significantly improved the prediction and accuracy of landcover classification. In this study, landcover classification is performed on seven-band Landsat 9 imagery and eight-band PlanetScope imagery for the village of Tully, NY, with an area of 900 square kilometers. The resolution of Landsat imagery is 30 meters, whereas the resolution of PlanetScope imagery is 3 meters. Classification schema is developed in ArcGIS Pro with five classification levels: conifer forest, hardwood forest, agriculture, developed, and water. Pixel-based supervised classification is performed using Support Vector Machine (SVM), Random Tress (RT), K-Nearest Neighbor (K-NN), and Maximum Likelihood Classifier (MLC). The reference dataset is acquired by an image interpreter using high-resolution imagery for map accuracy assessment. All the classification methods for Landsat imagery have more than 78% accuracy, but SVM performed best with 82% accuracy. For PlanetScope imagery, SVM performed best with 85% accuracy, whereas MLC had the lowest accuracy of 77%.

AGTML: A novel approach to land cover classification by integrating automatic generation of training samples and machine learning algorithms on Google Earth Engine

The timely, accurate, and automatic acquisition of land cover (LC) information is a prerequisite for detecting LC dynamics and performing ecological analyses. Cloud computing platforms, such as the Google Earth Engine, have substantially improved the efficiency and scale of LC classification. However, the lack of sufficient and representative training samples hinders automatic and accurate LC classification. In this study, we propose a new approach that integrates the automatic generation of training samples and machine learning algorithms (AGTML) for LC classification in Heilongjiang Province, China. After optimal focal radii were determined for different LC types using Landsat 8 based on focal statistics and unique phenology. Then target training samples were automatically generated based on the improved distance measure SED (a composite of Spectral angle distance (SAD) and Euclidean distance (ED)). Furthermore, LC classification was performed using four feature combinations and three machine learning algorithms. According to independent validation data, the automatically generated training samples demonstrated good representativeness and stability among all three classifiers, with an overall accuracy (OA) of classification higher than 86%, and showed high consistency in the landscape pattern of classification. RF yielded the highest classification accuracy (92.99% OA). AGTML outperformed GLC-FCS30 in identifying large fragmentation and small patch regions in the landscape types. The AGTML approach was subsequently applied to the Guanzhong Plain using different satellite imagery. Results were consistent and accurate (>96.50% OA), demonstrating that the AGTML approach can be applied to various regions and sensors, and has immense potential for automated LC classification across regional and global scales.

Local climate zone mapping using remote sensing: a synergetic use of daytime multi-view Ziyuan-3 stereo imageries and Luojia-1 nighttime light data

ABSTRACT The local climate zone (LCZ) scheme has been widely utilized in regional climate modeling, urban planning, and thermal comfort investigations. However, existing LCZ classification methods face challenges in characterizing complex urban structures and human activities involving local climatic environments. In this study, we proposed a novel LCZ mapping method that fully uses space-borne multi-view and diurnal observations, i.e. daytime Ziyuan-3 stereo imageries (2.1 m) and Luojia-1 nighttime light (NTL) data (130 m). Firstly, we performed land cover classification using multiple machine learning methods (i.e. random forest (RF) and XGBoost algorithms) and various features (i.e. spectral, textural, multi-view features, 3D urban structure parameters (USPs), and NTL). In addition, we developed a set of new cumulative elevation indexes to improve building roughness assessments. The indexes can estimate building roughness directly from fused point clouds generated by both along- and across-track modes. Finally, based on the land cover and building roughness results, we extracted 2D and 3D USPs for different land covers and used multi-classifiers to perform LCZ mapping. The results for Beijing, China, show that our method yielded satisfactory accuracy for LCZ mapping, with an overall accuracy (OA) of 90.46%. The overall accuracy of land cover classification using 3D USPs generated from both along- and across-track modes increased by 4.66%, compared to that of using the single along-track mode. Additionally, the OA value of LCZ mapping using 2D and 3D USPs (88.18%) achieved a better result than using only 2D USPs (83.83%). The use of NTL data increased the classification accuracy of LCZs E (bare rock or paved) and F (bare soil or sand) by 6.54% and 3.94%, respectively. The refined LCZ classification achieved through this study will not only contribute to more accurate regional climate modeling but also provide valuable guidance for urban planning initiatives aimed at enhancing thermal comfort and overall livability in urban areas. Ultimately, this study paves the way for more comprehensive and effective strategies in addressing the challenges posed by urban microclimates.

Improving the Accuracy of Land Use and Land Cover Classification of Landsat Data in an Agricultural Watershed

Classification of remotely sensed imagery for reliable land use and land cover (LULC) remains a challenge in areas where spectrally similar LULC features occur. For example, bare soils of harvested crop fields in agricultural watersheds exhibit spectral characteristics similar to high-intensity developed regions and impede an accurate classification. The goal of this study is to improve the accuracy of LULC classification of satellite imagery for the Big Sunflower River Watershed, Mississippi using ancillary data, multiple classification methods, and a post-classification correction (PCC). To determine the best approach, the methodology was applied to Landsat 8 Operational Land Imager (OLI) imagery during the growing season and post-harvest. Imagery for the growing season was acquired on 25 August 2015, and post-harvest was acquired on 7 January 2018. Three classification methods were applied: maximum likelihood (ML), support vector machine (SVM), and random forest (RF). LULC imagery was classified as open water, woody wetlands, harvested crop, rangeland, cultivated crop, high-intensity developed, and mid-low intensity developed areas. Ancillary data such as normalized difference vegetation index (NDVI), thematic maps of urban areas, river networks, transportation networks, high-resolution National Agriculture Imagery Program (NAIP) imagery, Google Earth time-series data, and phenology were used to determine the training dataset. Initially none of the three classification methods performed adequately. Hence, a post-classification correction (PCC) was implemented by masking and applying a majority filter using thematic maps of urban areas. Once PCC was implemented, the accuracies from each of the classification methods increased significantly with the SVM classification method performing best in both the growing season and post-harvest with an overall classification accuracy of 93.5% with a Kappa statistic of 0.88 in the post-harvest imagery and an overall classification accuracy of 84% with a Kappa statistic of 0.789 in the imagery from the growing season. It was found that SVM was the best classification method while PCC is an effective strategy to implement when dealing with spectrally similar LULC features. The use of SVM together with PCC increased the reliability of the information extracted. Strategies from this study can help to evaluate the LULC in agricultural and other watersheds.

The impact of selection of reference samples and DEM on the accuracy of land cover classification based on Sentinel-2 data

Up-to-date and reliable information on land cover and land use status is important in many aspects of human activities. Knowledge about the reference dataset, its coverage, nomenclature, thematic and geometric accuracy, spatial resolution is crucial for appropriate selection of reference samples used in the classification process. In this study, we examined the impact of the selection and pre-processing of reference samples for the classification accuracy. The classification based on Random Forest algorithm was performed using firstly the automatically selected reference samples derived directly from the national databases, and secondly using the pre-processed and verified reference samples. The verification procedures involved the iterative analysis of histogram of spectral features derived from the Sentinel-2 data for individual land cover classes. The verification of the reference samples improved the accuracy of delineation of all land cover classes. The highest improvement was achieved for the woodland broadleaved and non- and sparce vegetation classes, with the overall accuracy increasing from 51% to 73%, and from 33% to 74%, respectively. The second objective of this study was to derive the best possible land cover classification over the mountain area in Norway, therefore we examined whether the use of the Digital Elevation Model (DEM) can improve the classification results. Classifications were carried out based on Sentinel-2 data and a combination of Sentinel-2 and DEM. Using the DEM the accuracy for nine out of ten land cover classes was improved. The highest improvement was achieved for classes located at higher altitudes: low vegetation and non- and sparse vegetation.

Analysis of Land Cover Changes in the Brown Canyon Mining Area Associated with Restrictions on Community Activities

The high land use as area mining has resulted in changes in land cover and impacts the surrounding environment. In 2020, the Semarang City government implemented a Restrictions on Community Activities (RCA) policy to prevent the spread of the COVID-19 virus, indirectly limiting mining activities in the area of the Brown Canyon mine. This study aims to analyze the land cover change in the Brown Canyon mine area linked to the RCA policy. Assessment of land cover change is carried out by land cover classification on multitemporal Landsat imagery using Random Forest (RF) method. Based on the classification results, it was found that there was a change in mining land cover in Brown Canyon which increased by 3.713 Ha when the RCA policy was implemented from 46.305 Ha to 50.018 Ha. Likewise, after the implementation of RCA, there was an increase of 3.510 Ha from 50.018 Ha to 53.528 Ha. The accuracy of land cover classification before RCA was 90.12%. During RCA was 88.18%, and after RCA was 89.94%. Then, if related to the results of the interview data processing, the change in mining land cover that occurred was influenced by mining activities in the study area during the RCA period. So, it can also be said that RCA does not significantly influence mining activities in the Brown Canyon mining area. Several suggestions can be used for further research, such as using high-resolution satellite imagery and having mining area boundary data for each site.

Scale Effect of Land Cover Classification from Multi-Resolution Satellite Remote Sensing Data.

Land cover data are important basic data for earth system science and other fields. Multi-source remote sensing images have become the main data source for land cover classification. There are still many uncertainties in the scale effect of image spatial resolution on land cover classification. Since it is difficult to obtain multiple spatial resolution remote sensing images of the same area at the same time, the main current method to study the scale effect of land cover classification is to use the same image resampled to different resolutions, however errors in the resampling process lead to uncertainty in the accuracy of land cover classification. To study the land cover classification scale effect of different spatial resolutions of multi-source remote sensing data, we selected 1 m and 4 m of GF-2, 6 m of SPOT-6, 10 m of Sentinel-2, and 30 m of Landsat-8 multi-sensor data, and explored the scale effect of image spatial resolution on land cover classification from two aspects of mixed image element decomposition and spatial heterogeneity. For the study area, we compared the classification obtained from GF-2, SPOT-6, Sentinel-2, and Landsat-8 images at different spatial resolutions based on GBDT and RF. The results show that (1) GF-2 and SPOT-6 had the best classification results, and the optimal scale based on this classification accuracy was 4-6 m; (2) the optimal scale based on linear decomposition depended on the study area; (3) the optimal scale of land cover was related to spatial heterogeneity, i.e., the more fragmented and complex was the space, the smaller the scale needed; and (4) the resampled images were not sensitive to scale and increased the uncertainty of the classification. These findings have implications for land cover classification and optimal scale selection, scale effects, and landscape ecology uncertainty studies.