Approach For Land Cover Mapping Research Articles

Time-first approach for land cover mapping using big Earth observation data time-series in a data cube – a case study from the Lake Geneva region (Switzerland)

ABSTRACT Accurate, consistent, and high-resolution Land Use & Cover (LUC) information is fundamental for effectively monitoring landscape dynamics and better apprehending drivers, pressures, state, and impacts on land systems. Nevertheless, the availability of such national products with high thematic accuracy is still limited and consequently researchers and policymakers are constrained to work with data that do not necessarily reflect on-the-ground realities impending to correctly capture details of landscape features as well as limiting the identification and quantification of drivers and rate of change. Hereafter, we took advantage of the Switzerland’s official LUC statistical sampling survey and dense time-series of Sentinel-2 data, combining them with Machine and Deep Learning methods to produce an accurate and high spatial resolution land cover map over the Lake Geneva region. Findings suggest that time-first approach is a valuable alternative to space-first approaches, accounting for the intra-annual variability of classes, hence improving classification performances. Results demonstrate that Deep Learning methods outperform more traditional Machine Learning ones such as Random Forest, providing more accurate predictions with lower uncertainty. The produced land cover map has a high accuracy, an improved spatial resolution, while at the same time preserving the statistical significance (i.e. class proportion) of the official national dataset. This work paves the way towards the objective to produce a yearly high resolution land cover map of Switzerland and potentially implement a continuous land change monitoring capability. However further work is required to properly address challenges such as the need for increased temporal resolution for LUC information or the quality of training samples.

Weakly supervised high spatial resolution land cover mapping based on self-training with weighted pseudo-labels

Despite its success, deep learning in land cover mapping requires a massive amount of pixel-wise labeled images. It typically assumes that the training and test scenes are similar in data distribution. The performance of models trained on any particular dataset could degrade significantly on a new dataset due to the domain shift or domain gap across datasets, resulting in new training data requiring labor-intensive manual pixel-wise labeling. This paper proposes a land cover mapping framework combining Feature Pyramid Network (FPN) and self-training. In the FPN, we integrate ConvNeXt with a Pyramid Pooling Module (PPM). Combining the FPN and the PPM improves the segmentation performance, which benefits from the multiscale aggregation of pyramid features. To fully exploit pseudo-labels, we design an Unsupervised Domain Adaptation (UDA) land cover mapping scheme with self-training using weighted pseudo-labels of the target samples. The proposed land cover mapping framework could benefit from multiscale aggregation of pyramid features and the full use of the pseudo-labels. Comparison results on the LoveDA dataset, the latest large-scale unsupervised domain adaptation dataset for land cover mapping, empirically demonstrated that our land cover mapping approach significantly outperforms the baselines in both UDA scenarios, i.e., Urban → Rural and Rural → Urban . The models of this paper are now publicly available on GitHub. 1 1 https://github.com/csliujw/uda-self-training . • A land cover mapping network based on an FPN using ConvNeXt with a PPM to improve the land cover mapping performance. • A UDA scheme with self-training using weighted pseudo-labels. • Our scheme is significantly superior to the baselines on both UDA scenarios: Urban → Rural and Rural → Urban.

Mapping Urban and Peri-Urban Land Cover in Zimbabwe: Challenges and Opportunities

Accurate and current land cover information is required to develop strategies for sustainable development and to improve the quality of life in urban areas. This study presents an approach that combines multi-seasonal Sentinel-1 (S1) and Sentinel-2 (S2) data, and a random forest (RF) classifier in order to map land cover in four major urban centers in Zimbabwe. The specific objective of this study was to assess the potential of multi-seasonal (rainy, post-rainy, and dry season) S1, rainy season S2, post-rainy season, dry season S2, multi-seasonal S2, and multi-seasonal composite S1 and S2 data for mapping land cover in urban areas. The study results show that the combination of multi-seasonal S1 and S2 data improve land cover mapping in urban and peri-urban areas relative to only multi-seasonal S1, mono-seasonal S2, and multi-seasonal S2 data. The overall accuracy scores for the multi-seasonal S1 and S2 land cover maps are above 85% for all urban centers. Our results indicate that rainy and post-rainy S2 spectral bands, as well as dry-season S1 VV and VH bands (ascending orbit) are the most important features for land cover mapping. In particular, S1 data proved useful in separating built-up areas from cropland, which is usually problematic when only optical imagery is used in the study area. While there are notable improvements in land cover mapping, some challenges related to the S1 data analysis still remain. Nonetheless, our land cover mapping approach shows a potential to map land cover in other urban areas in Zimbabwe or in Sub-Sahara Africa. This is important given the urgent need for reliable geospatial information, which is required to implement the United Nations Sustainable Development Goals (UN SDGs) and United Nations New Urban Agenda (NUA) programmes.

An Cnn-Based Operational Approach For Land Cover Mapping in China

National land cover map with 30m resolution, an important database for studying the interaction between human and the environment, is a tedious work. The rise of deep learning technique provides a new idea for the work. This paper reports a novel method based on deep convolutional neural networks for the national land cover mapping task. The proposed method has four major parts: classification system, data sources, training samples selection, and training & inferencing. The produced deep learning (DL)-based land cover map is compared with two highly accepted land cover maps (the reference land cover map and the GLC30). Overall accuracies of GLC30 and DL-land cover map are 76.45% and 82.59% when considering the reference land cover map as the ground truth. Overall accuracies of the reference land cover map and the DL-land cover map are 74.25% and 78.87% when the GLC30 is treated as the ground truth.

OIC-MCE: A Practical Land Cover Mapping Approach for Limited Samples Based on Multiple Classifier Ensemble and Iterative Classification

Land cover samples are usually the foundation for supervised classification. Unfortunately, for land cover mapping in large areas, only limited samples can be used due to the time-consuming and labor-intensive sample collection. A novel and practical Object-oriented Iterative Classification method based on Multiple Classifiers Ensemble (OIC-MCE) was proposed in this paper. It systematically integrated object-oriented segmentation, Multiple Classifier Ensemble (MCE), and Iterative Classification (IC). In this method, the initial training samples were updated self-adaptively during the iterative processes. Based on these updated training samples, the inconsistent regions (ICR) in the classification results of the MCE method were reclassified to reduce their uncertainty. Three typical case studies in the China-Pakistan Economic Corridor (CPEC) indicate that the overall accuracy of the OIC-MCE method is significantly higher than that of the single classifier. After five iterations, the overall accuracy of the OIC-MCE approach increased by 5.58%–8.38% compared to the accuracy of the traditional MCE method. The spatial distribution of newly added training samples generated by the OIC-MCE approach was relatively uniform. It was confirmed by ten repeated experiments that the OIC-MCE approach has good stability. More importantly, even if the initial sample size reduced by 65%, the quality of the final classification result based on the proposed OIC-MCE approach would not be greatly affected. Therefore, the proposed OIC-MCE approach provides a new solution for land cover mapping with limited samples. Certainly, it is also well suited for land cover mapping with abundant samples.

A spatialized classification approach for land cover mapping using hyperspatial imagery

Maps of classified surface features are a key output from remote sensing. Conventional methods of pixel-based classification label each pixel independently by considering only a pixel's spectral properties. While these purely spectral-based techniques may be applicable to many medium and coarse-scale remote sensing analyses, they may become less appropriate when applied to high spatial resolution imagery in which the pixels are smaller than the objects to be classified. At this scale, there is often higher intra-class spectral heterogeneity than inter-class spectral heterogeneity, leading to difficulties in using purely spectral-based classifications. A solution to these issues is to use not only a pixel's spectral characteristics but also its spatial characteristics. In this study, we develop a generalizable "spatialized" classification approach for high spatial resolution image classification. We apply the proposed approach to map vegetation growth forms such as trees, shrubs, and herbs in a forested ecosystem in the Sierra Nevada Mountains. Our results found that the spatialized classification approach outperformed spectral-only approaches for all cover classes examined, with the largest improvements being in discriminating vegetation classes.

A SPECLib-based operational classification approach: A preliminary test on China land cover mapping at 30 m

Fine-resolution land cover mapping at the regional or global scale is usually time-consuming and involves a lot of manual participation. This article proposes a novel and automatic land cover mapping approach called the SPatial-tEmporal speCtral Library (SPECLib), which can produce a land cover map with a 30-m spatial resolution. First, the SPECLib is developed by extracting reference spectra for uniform, typical samples of each land cover type by combining the GlobCover2009 land cover product and the 8-day composite MODIS Surface Reflectance Product (MOD09 A1). SPECLib has a temporal resolution of 8 days and contains most of the land cover types in each 1.5°×1.5° geographic cell. For an arbitrary temporal stage, for each geographical cell of the SPECLib, there are a total of about 5000 MODIS spectra. Secondly, an operational approach for automatic land cover mapping using Landsat data is developed based on SPECLib. For each Landsat scene to be classified, the reference spectra of the land cover types in the corresponding geographic cell were automatically extracted and normalized. The Landsat scene was then classified by using the normalized reference spectra from SPECLib in an integrated classification approach that combined MLC (Maximum Likelihood Classification) and SAM (Spectral Angle Mapper). Finally, the SPECLib-based approach is tested using 510 single-date Landsat 8 OLI scenes to produce a land cover map of China. The map was validated using 11,366 interpretation samples and was shown to have an overall accuracy of 77.34% and a kappa coefficient of 0.729. Therefore, the preliminary results indicate that the SPECLib-based operational classification approach is a promisingly automatic method for global land cover mapping at a resolution of 30 m.

Towards automated 10–30 m resolution land cover mapping in insular South-East Asia

In this study, we create and critically analyse an automated decision tree classification approach for regional level land cover mapping in insular South-East Asian conditions, using a combination of 10–30 m resolution optical and radar data. The resulting map contains 11 land cover classes and reveals a great deal of contextual information due to high spatial resolution. A limited accuracy assessment indicates 59–97% class wise accuracies. The unprecedented spatial detail of closed canopy oil palm mapping (with user’s accuracy of 90%) is seen as the most promising feature of the mapping approach. The incapability of separating primary forests from other tree cover, and the large variety of different landscapes (e.g. home gardens and tea plantations) classified as shrubland, are considered the main areas for future improvement. Overall, the study demonstrates the great potential of multi-source 10–30 m resolution high data volume land cover mapping approaches in insular South-East Asian conditions.

Land cover mapping in cropland dominated area using information on vegetation phenology and multi-seasonal Landsat 8 images

Information about land cover is required for economic, agricultural and environmental policy making. Therefore, reliable up-to-date information is always called upon. In this study, we developed a new approach for land cover mapping based on the information of vegetation phenology. The main objective of this approach was to generate a land cover map of large cropland dominated area with high classification accuracy. Our approach consisted of two steps: first, we divided the study area into three land use groups depending on the phenology trend of cereals. Second, we applied a supervised classification for each group using the Maximum Likelihood Classifier and multi-date satellite images. Recent multi-temporal Landsat 8 images and field survey data were used for the classification process. To assess the robustness of this approach, a conventional supervised classification was performed using single date and multi-date images. Results indicated that the proposed approach is able to discriminate between different land cover types which have a similar spectral reflectance such as cereals, vegetables and pasture with high accuracy. The accuracy assessment showed very promising results with an overall accuracy of 86 % and a Kappa of 0.85 (good agreement) as compared to the single date (54–55 %) and the multi-date approach (78 %). Indeed, the application of this method provides accurate information for ecologists, hydrologists and the land development decision-makers. It can also improve the accuracy of environmental models that require high resolution land cover maps as input data.

Comparative analysis of product-level fusion, support vector machine, and artificial neural network approaches for land cover mapping

Increasing the accuracy of thematic maps generated using satellite imagery is a crucial task in remote sensing. In this study, a product-level fusion (PLF) approach based on integration of different land-type maps generated using various satellite-derived indices including normalized difference water index (NDWI), normalized difference built-up index (NDBI), enhanced vegetation index (EVI), and normalized difference vegetation index (NDVI) is proposed to improve the accuracy of land cover mapping. The suitability of the proposed approach for land cover mapping is evaluated in comparison with two high-performance image classification techniques including support vector machine (SVM) and artificial neural network (ANN). The results show that the overall accuracy and kappa values of about 95.95 % and 0.95, 94.91 % and 0.94, and 85.32 % and 0.82 are achieved for the PLF, SVM, and ANN approaches, respectively. The results indicate superiority of the PLF approach than SVM and ANN techniques for land cover classification of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) imagery, especially for the extraction of forest, rice, and citrus classes. However, SVM technique also provided reliable result for land cover mapping.

Can the Future EnMAP Mission Contribute to Urban Applications? A Literature Survey

With urban populations and their footprints growing globally, the need to assess the dynamics of the urban environment increases. Remote sensing is one approach that can analyze these developments quantitatively with respect to spatially and temporally large scale changes. With the 2015 launch of the spaceborne EnMAP mission, a new hyperspectral sensor with high signal-to-noise ratio at medium spatial resolution, and a 21 day global revisit capability will become available. This paper presents the results of a literature survey on existing applications and image analysis techniques in the context of urban remote sensing in order to identify and outline potential contributions of the future EnMAP mission. Regarding urban applications, four frequently addressed topics have been identified: urban development and planning, urban growth assessment, risk and vulnerability assessment and urban climate. The requirements of four application fields and associated image processing techniques used to retrieve desired parameters and create geo-information products have been reviewed. As a result, we identified promising research directions enabling the use of EnMAP for urban studies. First and foremost, research is required to analyze the spectral information content of an EnMAP pixel used to support material-based land cover mapping approaches. This information can subsequently be used to improve urban indicators, such as imperviousness. Second, we identified the global monitoring of urban areas as a promising field of investigation taking advantage of EnMAP’s spatial coverage and revisit capability. However, owing to the limitations of EnMAPs spatial resolution for urban applications, research should also focus on hyperspectral resolution enhancement to enable retrieving material information on sub-pixel level.

Association of West Nile virus illness and urban landscapes in Chicago and Detroit

BackgroundWest Nile virus infection in humans in urban areas of the Midwestern United States has exhibited strong spatial clustering during epidemic years. We derived urban landscape classes from the physical and socio-economic factors hypothesized to be associated with West Nile Virus (WNV) transmission and compared those to human cases of illness in 2002 in Chicago and Detroit. The objectives were to improve understanding of human exposure to virus-infected mosquitoes in the urban context, and to assess the degree to which environmental factors found to be important in Chicago were also found in Detroit.ResultsFive urban classes that partitioned the urban space were developed for each city region. The classes had many similarities in the two settings. In both regions, the WNV case rate was considerably higher in the urban class associated with the Inner Suburbs, where 1940–1960 era housing dominates, vegetation cover is moderate, and population density is moderate. The land cover mapping approach played an important role in the successful and consistent classification of the urban areas.ConclusionThe analysis demonstrates how urban form and past land use decisions can influence transmission of a vector-borne virus. In addition, the results are helpful to develop hypotheses regarding urban landscape features and WNV transmission, they provide a structured method to stratify the urban areas to locate representative field study sites specifically for WNV, and this analysis contributes to the question of how the urban environment affects human health.

Land cover assessment with MODIS imagery in southern African Miombo ecosystems

Global land use and land cover products in highly dynamic tropical ecosystems lack the detail needed for natural resource management and monitoring at the national and provincial level. The MODIS sensor provides improved opportunities to combine multispectral and multitemporal data for land use and land cover mapping. In this paper we compare the MODIS Global Land Cover Classification Product with recent land use and land cover maps at the national level over a characteristic location of Miombo woodlands in the province of Zambezia, Mozambique. The performances of three land cover-mapping approaches were assessed: single-date supervised classification, principal component analysis of band-pair difference images, and multitemporal NDVI analysis. Extensive recent field data were used for the definition of the test sites and accuracy assessment. Encouraging results were achieved with the three approaches. The classification results were refined with the help of a digital elevation model. The most consistent results were achieved using principal component analysis of band-pair difference images. This method provided the most accurate classifications for agriculture, wetlands, grasslands, thicket and open forest. The overall classification accuracy reached 90%. The multitemporal NDVI provided a more accurate classification for the dense forest cover class. The selection of the right image dates proved to be critical for all the cases evaluated. The flexibility of these alternatives makes them promising options for rapid and inexpensive land cover mapping in regions of high environmental variability such as tropical developing countries.

Land cover mapping principles: A return to interpretation fundamentals

Land cover mapping methodology should be governed by the attributes of the land cover types. Since land cover types usually have significantly different interpretation characteristics, they should be mapped separately according to the methodology suitable to their particular attributes. Land cover mapping requires the use of other interpretation elements than just reflectance, particularly position in the landscape, size and association. In addition, knowledge of the land use systems and their extent is essential for attributing spectral reflectance curves to land cover types. A bottom-up land cover mapping approach, using classical interpretation principles, but facilitated with modern remote sensing software, is recommended. The procedure has been applied to mapping land cover types of southern Sumatra from Landsat and Spot imagery.