Exploring the synergistic use of multi-scale image object metrics for land-use/land-cover mapping using an object-based approach

This study presents a new method for the synergistic use of multi-scale image object metrics for land-use/land-cover mapping using an object-based classification approach. This new method can integrate an object with its super-objects’ metrics. The entire classification involves two object hierarchies: (1) a five-level object hierarchy to extract object metrics at five scales, and (2) a three-level object hierarchy for the classification process. A five-level object hierarchy was developed through multi-scale segmentation to calculate and extract both spectral and textural metrics. Layers representing the hierarchy at each of the five scales were then intersected by using the overlay tool, an intersected layer was created with metrics from all five scales, and the same geometric elements were retained as those of the objects of the lowest level. A decision tree analysis was then used to build rules for the classification of the intersected layer, which subsequently served as the thematic layer in a three-level object hierarchy to identify shadow regions and produce the final map. The use of multi-scale object metrics yielded improved classification results compared with single-scale metrics, which indicates that multi-scale object metrics provide valuable spatial information. This method can fully utilize metrics at multiple scales and shows promise for use in object-based classification approaches.     

A quantitative investigation of the uncertainty associated with mapping scale in the production of land-cover/land-use data

ABSTRACT

Mapping scale is an essential issue in land use and land cover (LULC) data production, which always involves the minimum mapping unit (MMU) that stipulated in the product specification. Since the application of MMUs will inevitably cause some inappropriate classification problems, a technique is needed to evaluate the impact on the data outputs. In this study, a novel method is proposed to investigate the classification uncertainty brought by MMUs on LULC data. The omission errors are predicted based on an assumption of the skewed frequency distribution of the LULC patch size, and the commission errors are subsequently computed through the conversion possibilities among different land classes, which can be deduced from the generalization rule. A test is conducted on real data to verify the underlying assumption on the patch size distribution, and the accuracy of the prediction of omission errors is evaluated through a simulation experiment. A case study is also presented to demonstrate the efficiency and feasibility of the proposed method. At the end of this article, the advantages and notes of this method are discussed for further study and application.     

Coupling spectral unmixing and multiseasonal remote sensing for temperate dryland land-use/land-cover mapping in Minqin County,China

Temperate dryland ecosystems in China are undergoing accelerated changes due to natural and anthropogenic disturbances. Using the Minqin oasis as a case study area, this article examined linear spectral mixture analysis (LSMA) with the fixed four-image endmember (EM) model comprising sand, green vegetation, saline land, and dark materials (i.e. the degraded symptoms/EMs) for temperate dryland land-use and land-cover mapping. Dryland covers defined by landscape seasonality of the four EMs at the subpixel level are more easily interpreted to achieve acceptable accuracy, and allow better understanding of the processes of land degradation, such as two distinct salt/water movements were found in the study area. The Minqin oasis faces a significant challenge that requires a long-term monitoring system to understand the relationship between land-use decisions and ecological consequences. The approach developed with the mutiseasonal LSMA, representing dryland land-cover seasonality with the common surface degradation types (i.e. the four EMs) in a tree-structure framework, promises a robust and operative tool for land degradation assessment and monitoring and would be applied in a more different dryland environment in the future.     

Using MERIS fused images for land-cover mapping and vegetation status assessment in heterogeneous landscapes

In this paper we evaluate the potential of ENVISAT–Medium Resolution Imaging Spectrometer (MERIS) fused images for land-cover mapping and vegetation status assessment in heterogeneous landscapes. A series of MERIS fused images (15 spectral bands; 25 m pixel size) is created using the linear mixing model and a Landsat Thematic Mapper (TM) image acquired over the Netherlands. First, the fused images are classified to produce a map of the eight main land-cover types of the Netherlands. Subsequently, the maps are validated using the Dutch land-cover/land-use database as a reference. Then, the fused image with the highest overall classification accuracy is selected as the best fused image. Finally, the best fused image is used to compute three vegetation indices: the normalized difference vegetation index (NDVI) and two indices specifically designed to monitor vegetation status using MERIS data: the MERIS terrestrial chlorophyll index (MTCI) and the MERIS global vegetation index (MGVI).

Results indicate that the selected data fusion approach is able to downscale MERIS data to a Landsat-like spatial resolution. The spectral information in the fused images originates fully from MERIS and is not influenced by the TM data. Classification results for the TM and for the best fused image are similar and, when comparing spectrally similar images (i.e. TM with no short-wave infrared bands), the results of the fused image outperform those of TM. With respect to the vegetation indices, a good correlation was found between the NDVI computed from TM and from the best fused image (in spite of the spectral differences between these two sensors). In addition, results show the potential of using MERIS vegetation indices computed from fused images to monitor individual fields. This is not possible using the original MERIS full resolution image. Therefore, we conclude that MERIS–TM fused images are very useful to map heterogeneous landscapes.     

An overview of 21 global and 43 regional land-cover mapping products

Land-cover (LC) products, especially at the regional and global scales, comprise essential data for a wide range of environmental studies affecting biodiversity, climate, and human health. This review builds on previous compartmentalized efforts by summarizing 23 global and 41 regional LC products. Characteristics related to spatial resolution, overall accuracy, time of data acquisition, sensor used, classification scheme and method, support for LC change detection, download location, and key corresponding references are provided. Operational limitations and uncertainties are discussed, mostly as a result of different original modelling outcomes. Upcoming products are presented and future prospects towards increasing usability of different LC products are offered. Despite the common realization of product usage by non-experts, the remote-sensing community has not fully addressed the challenge. Algorithmic development for the effective representation of inherent product limitations to facilitate proper usage by non-experts is necessary. Further emphasis should be placed on international coordination and harmonization initiatives for compatible LC product generation. We expect the applicability of current and future LC products to increase, especially as our environmental understanding increases through multi-temporal studies.     

Spatiotemporal analysis of land-use and land-cover change in the Brazilian Amazon

This paper provides a comparative analysis of land-use and land-cover (LULC) changes among three study areas with different biophysical environments in the Brazilian Amazon at multiple scales, from per-pixel, polygon, census sector, to study area. Landsat images acquired during the years of 1990/1991, 1999/2000, and 2008/2010 were used to examine LULC change trajectories with the post-classification comparison approach. A classification system composed of six classes – forest, savanna, other vegetation (secondary succession and plantations), agro-pasture, impervious surface, and water – was designed for this study. A hierarchical-based classification method was used to classify Landsat images into thematic maps. This research shows different spatiotemporal change patterns, composition, and rates among the three study areas and indicates the importance of analysing LULC change at multiple scales. The LULC change analysis over time for entire study areas provides an overall picture of change trends, but detailed change trajectories and their spatial distributions can be better examined at a per-pixel scale. The LULC change at the polygon scale provides the information of the changes in patch sizes over time, while the LULC change at census sector scale gives new insights on how human-induced activities (e.g. urban expansion, roads, and land-use history) affect LULC change patterns and rates. This research indicates the necessity to implement change detection at multiple scales for better understanding the mechanisms of LULC change patterns and rates.     

Towards a common validation sample set for global land-cover mapping

Validating land-cover maps at the global scale is a significant challenge. We built a global validation data-set based on interpreting Landsat Thematic Mapper (TM) and Enhanced TM Plus (ETM+) images for a total of 38,664 sample units pre-determined with an equal-area stratified sampling scheme. This was supplemented by MODIS enhanced vegetation index (EVI) time series data and other high-resolution imagery on Google Earth. Initially designed for validating 30 m-resolution global land-cover maps in the Finer Resolution Observation and Monitoring of Global Land Cover (FROM-GLC) project, the data-set has been carefully improved through several rounds of interpretation and verification by different image interpreters, and checked by one quality controller. Independent test interpretation indicated that the quality control correctness level reached 90% at level 1 classes using selected interpretation keys from various parts of the USA. Fifty-nine per cent of the samples have been verified with high-resolution images on Google Earth. Uncertainty in interpretation was measured by the interpreter’s perceived confidence. Only less than 7% of the sample was perceived as low confidence at level 1 by interpreters. Nearly 42% of the sample units located within a homogeneous area could be applied to validating global land-cover maps whose resolution is 500 m or finer. Forty-six per cent of the sample whose EVI values are high or with little seasonal variation throughout the year can be applied to validate land-cover products produced from data acquired in different phenological stages, while approximately 76% of the remaining sample whose EVI values have obvious seasonal variation was interpreted from images acquired within the growing season. While the improvement is under way, some of the homogeneous sample units in the data-set have already been used in assessing other classification results or as training data for land-cover mapping with coarser-resolution data.     

Analysis of L-band SAR backscatter and coherence for delineation of land-use/land-cover

In this study, we investigated the potential improvement of land-use/land-cover (LU/LC) classification using multidate backscatter intensity as well as interferometric coherence images derived from Advanced Land Observing Satellite phased array L-band synthetic aperture radar data. Four interferometric synthetic aperture radar data pairs in horizontal–horizontal polarizations were processed to obtain backscatter intensity and coherence images. From the analysis of these images, it was observed that backscatter values alone are not sufficient to separate certain LU/LC classes, e.g. forest and mining areas, due to similarities in the associated scattering mechanisms producing similar backscatter values. However, the temporal coherence values from these LU/LC features were found to be distinctly different. Supervised classifications using maximum-likelihood distance were performed with various combinations of data (three-date backscatter intensity and two-date backscatter intensity with corresponding coherence data) to generate LU/LC maps of the study area. The comparison of classification accuracies obtained for different combinations of data indicates that the classification accuracy is improved by adding coherence information to the backscatter intensity data compared to using the multidate backscatter intensity data alone. Thus, the analysis of backscatter intensity along with coherence is a better alternative than using backscatter intensity alone to improve the accuracy in LU/LC classification.     

An all-season sample database for improving land-cover mapping of Africa with two classification schemes

High-quality training and validation samples are critical components of land-cover and land-use mapping tasks in remote sensing. For large area mapping it is much more difficult to build such sample sets due to the huge amount of work involved in sample collection and image processing. As more and more satellite data become available, a new trend emerges in land-cover mapping that takes advantage of images acquired beyond the greenest season. This has created the need for constructing sample sets that can be used in classifying images of multiple seasons. On the other hand, seasonal land-cover information is also becoming a new demand in land and climate change studies. Here we produce the first training and validation data sets with seasonal labels in order to support the production of seasonal land-cover data for entire Africa. Nonetheless, for the first time, two classification systems were created for the same set of samples. We adapted the finer resolution observation and monitoring of global land cover (FROM-GLC) and the Food and Agriculture Organization (FAO) Land Cover Classification System legends. Locations of training-sample units of FROM-GLC were repurposed here. Then we designed a process to enlarge the training-sample units to increase the density of samples in the feature space of spectral characteristics of Moderate Resolution Imaging Spectroradiometer (MODIS) time-series and Landsat imagery. Finally, we obtained 15,799 training-sample units and 7430 validation-sample units. The land-cover type at each point was recorded at the time of maximum greenness in addition to four seasons in a year. Nearly half of the sample units were also suitable for 500 m resolution MODIS data. We analysed the representativeness of the training and validation sets and then provided some suggestions about their use in improving classification accuracies of Africa.     

The application of ensemble techniques for land-cover classification in arid lands

Robust classification approaches are required for accurate classification of complex land-use/land-cover categories of desert landscapes using remotely sensed data. Machine-learning ensemble classifiers have proved to be powerful for the classification of remotely sensed data. However, they have not been evaluated for classifying land-cover categories in desert regions. In this study, the performance of two machine-learning ensemble classifiers – random forests (RF) and boosted artificial neural networks – is explored in the context of classification of land use/land cover of desert landscapes. The evaluation is based on the accuracy of classification of remotely sensed data, with and without integration of ancillary data. Landsat-5 Thematic Mapper data captured for a desert landscape in the north-western coastal desert of Egypt are used with ancillary variables derived from a digital terrain model to classify 13 different land-use/land-cover categories. Results show that the two ensemble methods produce accurate land-cover classifications, with and without integrating spectral data with ancillary data. In general, the overall accuracy exceeded 85% and the kappa coefficient (κ) attained values over 0.83. The integration of ancillary data improved the performance of the boosted artificial neural networks by approximately 5% and the random forests by 9%. The latter showed overall higher accuracy; however, boosted artificial neural networks showed better generalization ability and lower overfitting tendencies. The results reveal the merit of applying ensemble methods to integrated spectral and ancillary data of similar desert landscapes for achieving high classification accuracies.     

Multi-resolution and temporal characterization of land-use classes in a Mediterranean wetland with land-cover fractions

Four different methods for analysing land-use and land-cover fractions at multiple scales, namely composite operator, t-test, Dutilleul’s modified t-test and ternary diagrams of physical models for process pathways, were applied to sets of multi-resolution images in order to evaluate the usefulness of coarse-resolution satellite data (e.g. the Moderate Resolution Imaging Spectroradiometer; MODIS) in obtaining similar results to those obtainable with moderate-resolution satellite data (e.g. Landsat). A spectral-mixture model based on three endmembers (soil, vegetation and water) was used to determine the land-cover fractions of the main land-use classes of a wetland in southeast Spain. The land-use map was produced by applying the unsupervised k-means classification method to the moderate-resolution image. Spatial and temporal changes in the mixture fractions at multiple resolutions and their corresponding land-cover fraction maps were assessed. Three different t-tests (paired-samples, independent-samples and Dutilleul’s modified t-tests) were used to evaluate the effects of pixel aggregation on land-cover fractions and land-use maps in terms of surface-area estimations. Ternary plots of land-use classes characterized by land-cover fractions were used to visualize environmental processes pathways describing temporal changes in the landscape. The results obtained with moderate- and coarse-resolution data were not significantly different from each other. Land-use and land-cover surface-area estimations were not significantly different between Landsat moderate-resolution (30 m) and Landsat resampled coarse-resolution (300 m) data. Spatial autocorrelation had an important effect when comparing Landsat moderate-resolution (30 m) with MODIS coarse-resolution (250 m) data. In order to minimize these effects Dutilleul’s modified t-test was applied for the comparison of Landsat with MODIS image data. However, this test did not reveal significant differences between both datasets, whereas with the ordinary t-test, significant differences were found, which suggest the existence of a bias by spatial autocorrelation that must be taken into account for up-scaling or down-scaling of remote-sensing data. The results suggest the possibility of using coarse-resolution images (MODIS) to characterize environmental changes with a similar accuracy to those of moderate-resolution images (Landsat), as long as potential spatial autocorrelation effects are taken into account. This finding indicates that a substantial reduction in the costs of conducting wetland management and monitoring tasks can be achieved by using free or low-cost coarse-resolution satellite images.     

Application of spectral mixture analysis to Amazonian land-use and land-cover classification

Abundant vegetation species and associated complex forest stand structures in moist tropical regions often create difficulties in accurately classifying land-use and land-cover (LULC) features. This paper examines the value of spectral mixture analysis (SMA) using Landsat Thematic Mapper (TM) data for improving LULC classification accuracy in a moist tropical area in Rondônia, Brazil. Different routines, such as constrained and unconstrained least-squares solutions, different numbers of endmembers, and minimum noise fraction transformation, were examined while implementing the SMA approach. A maximum likelihood classifier was also used to classify fraction images into seven LULC classes: mature forest, intermediate secondary succession, initial secondary succession, pasture, agricultural land, water, and bare land. The results of this study indicate that reducing correlation between image bands and using four endmembers improve classification accuracy. The overall classification accuracy was 86.6% for the seven LULC classes using the best SMA processing routine, which represents very good results for such a complex environment. The overall classification accuracy using a maximum likelihood approach was 81.4%. Another finding is that use of constrained or unconstrained solutions for unmixing the atmospherically corrected or raw Landsat TM images does not have significant influence on LULC classification performances when image endmembers are used in a SMA approach.     

A refined classification approach by integrating Landsat Operational Land Imager (OLI) and RADARSAT-2 imagery for land-use and land-cover mapping in a tropical area

Producing accurate land-use and land-cover (LULC) mapping is a long-standing challenge using solely optical remote-sensing data, especially in tropical regions due to the presence of clouds. To supplement this, RADARSAT images can be useful in assisting LULC mapping. The fusion of optical and active remote-sensing data is important for accurate LULC mapping because the data from different parts of the spectrum provide complementary information and often lead to increased classification accuracy. Also, the timeliness of using synthetic aperture radar (SAR) fills information gaps during overcast or hazy periods. Therefore, this research designed a refined classification procedure for LULC mapping for tropical regions. Determining the best method for mapping with a specific data source and study area is a major challenge because of the wide range of classification algorithms and methodologies available. In this study, different combinations and the potential of Landsat Operational Land Imager (OLI) and RADARSAT-2 SAR data were evaluated to select the best procedure for LULC classification. Results showed that the best filter for SAR speckle reduction is the 5 × 5 enhanced Lee. Furthermore, image-sharpening algorithms were employed to fuse Landsat multispectral and panchromatic bands and subsequently these algorithms were analysed in detail. The findings also confirmed that Gram–Schmidt (GS) performed better than the other techniques employed. Fused Landsat data and SAR images were then integrated to produce the LULC map. Different classification algorithms were adopted to classify the integrated Landsat and SAR data, and the maximum likelihood classifier (MLC) was considered the best approach. Finally, a suitable classification procedure was designed and proposed for LULC as mapping in tropical regions based on the results obtained. An overall accuracy of 98.62% was achieved from the proposed methodology. The proposed methodology is a useful tool in industry for mapping purposes. Additionally, it is also useful for researchers, who could extend the method for different data sources and regions.     

An assessment of landsat MSS and TM data for urban and near-urban land-cover digital classification

The information content of Landsat TM and MSS data was examined to assess the ability to digitally differentiate urban and near-urban land covers around Miami, FL. This examination included comparisons of unsupervised signature extractions for various cover types, training site statistics for intraclass and interclass separability, and band and band combination selection from an 11-band multisensor data set. The principal analytical tool used in this study was transformed divergence calculations. The TM digital data are typically more useful than the MSS data in the homogeneous near-urban land-covers and less useful in the heterogeneous urban areas.     

An assessment of impacts of land-cover changes on root-zone soil moisture

Land cover exerts considerable control over the exchange of energy, water, and carbon dioxide and other greenhouse gases between land surface and the atmosphere. In China, dramatic land-cover changes have occurred along with rapid economic development in the past 30 years. However, research specifically on whether such land-cover changes have any influence on root-zone soil moisture in the region has started only in very recent few years. In this study, the performance of selected land-surface models (Noah 2.7.1, Noah 3.2, Common Land Model (CLM version 2.0), and Mosaic) implemented in National Aeronautics and Space Administration (NASA)’s Land Information System (LIS version 6.1.6) is first tested using quality-controlled soil moisture observations from 108 in situ sites of the China Meteorological Administration. The best-performing model (CLM2.0) is selected to estimate the influence of land-cover changes on root-zone soil moisture, as well as drought occurrence in Yunnan Province in China. Both the 1992–1993 Advanced Very High Resolution Radiometer (AVHRR) and 2007–2010 Moderate Resolution Imaging Spectroradiometer Collection 5 (MODIS) land-cover products at 1 km resolution are employed to represent 1990 and 2010 land-cover status, respectively. These are verified using the local ground records of Yunnan Province over the two time periods. Their differences are considered roughly as land-cover changes occurring during the period 1990–2010. It is found that land-cover changes from primeval forest to grassland may increase root-zone soil moisture, thus reducing drought, while changes from grassland and primeval forest to cropland or reforested areas have increased the likelihood of drought.     

A hybrid approach to mapping land-use modification and land-cover transition from MODIS time-series data: A case study from the Bolivian seasonal tropics

This study uses a combination of satellite imagery and GIS data, a vegetation map, interview data, and on-site field studies to map detailed natural vegetation to land-use conversion pathways (~ 22,000 possible combinations) in the seasonal tropics of Santa Cruz Department in southeastern Bolivia from 1994 to 2008. We mapped a suite of land-use classes based on the seasonal phenology of double- and single season cropping regimes; pasture; and bare soil cropland (fallow). Analyses focus specifically on the Corredor Bioceánico, which bisects some of the most sensitive and poorly understood ecosystems in the world and indirectly creating one of the most important agricultural region-deforestation hotspots in South America at the present time. Training data to predict class membership were based on MODIS NDVI annual mean, maximum, minimum, and amplitude derived from field observations, semi-structured interviews, and aerial videography. Results show that over 8,000 km² of forest was lost during the 14-year study period. In the first years of cultivation, pasture is the dominant land use, but quickly gives way to cropland. The main findings according to forest type is that transitional forest types on deep and poorly drained soils of alluvial plains have lost the most in terms of percentage area cleared. The resulting transition pathways can potentially provide decision-makers with more detailed insight as to the proximate causes or driving forces of land change in addition to the most threatened forests remaining in the Tierras Bajas and those most likely to be cleared in the Brazilian Shield and Pantanal.     

Classifier ensemble construction with rotation forest to improve medical diagnosis performance of machine learning algorithms

Improving accuracies of machine learning algorithms is vital in designing high performance computer-aided diagnosis (CADx) systems. Researches have shown that a base classifier performance might be enhanced by ensemble classification strategies. In this study, we construct rotation forest (RF) ensemble classifiers of 30 machine learning algorithms to evaluate their classification performances using Parkinson's, diabetes and heart diseases from literature.While making experiments, first the feature dimension of three datasets is reduced using correlation based feature selection (CFS) algorithm. Second, classification performances of 30 machine learning algorithms are calculated for three datasets. Third, 30 classifier ensembles are constructed based on RF algorithm to assess performances of respective classifiers with the same disease data. All the experiments are carried out with leave-one-out validation strategy and the performances of the 60 algorithms are evaluated using three metrics; classification accuracy (ACC), kappa error (KE) and area under the receiver operating characteristic (ROC) curve (AUC).Base classifiers succeeded 72.15%, 77.52% and 84.43% average accuracies for diabetes, heart and Parkinson's datasets, respectively. As for RF classifier ensembles, they produced average accuracies of 74.47%, 80.49% and 87.13% for respective diseases.RF, a newly proposed classifier ensemble algorithm, might be used to improve accuracy of miscellaneous machine learning algorithms to design advanced CADx systems.     

An assessment of land-use change in the Cocoa Belt of south-west Nigeria

The Nigerian government is reviving the agricultural sector to shift from its sole dependence on crude oil for foreign exchange earnings. Thus, the Cocoa Belt (agro-ecological region) of southwest Nigeria is important to the national economy. With the increasing demand for land to grow export crops and to meet other needs such as settlement expansion, land use is changing. Land-use data and mapping are essential inputs for the process of formulating, implementing, and monitoring policy with the aim of reducing the impact of land-cover/land-use (LCLU) change. Land-use types, their spatial extent and dynamics over a 25 year period are examined from multispectral images of the Landsat Thematic Mapper and Enhanced Thematic Mapper Plus. This study examines the main drivers of LCLU change and the environmental impact. Results show that forest conversion to agricultural lands is the main trend, and cultivation is the main cause of forest loss in the study area. The need to produce food for the teeming population, coupled with the government's policy to expand export crop production is resulting in the loss of native forest, including areas designated as forest reserves. Results underscore the need for deliberate land-use planning and management in this belt. This study reveals the situation of unplanned and rapid changes to land use in the context of a developing country where explicit policies to cater for such activities are absent.     

Multitemporal MERIS images for land-cover mapping at a national scale: a case study of Portugal

Medium-spatial-resolution satellite images have already proved to be successful in automatic production of global land-cover maps. However, their operational use for land-cover mapping at a national scale has not yet been well established. We find that the reasons for this are not data-source dependent, but are due to the land-cover nomenclatures properties adopted, regional landscape specificities and the methodological approaches used. The aim of this paper is to evaluate the suitability for national applications of land-cover maps derived from automatic classification of medium-spatial-resolution satellite images. To tackle this issue, we produce a land-cover map of Continental Portugal from multitemporal MEdium Resolution Imaging Spectrometer (MERIS) full-resolution satellite images of 2005 and evaluate its accuracy. For the accuracy assessment of the final map, we compute unbiased estimates of overall, user and producer accuracies using an independent testing sample collected through a stratified random sampling design. The overall accuracy of the final map is 80%, with an absolute precision of 2% at the 95% confidence level. High independent accuracy assessment results demonstrate that medium-spatial-resolution satellite images can be used on an operational basis for annual production of land-cover maps suitable for national applications.