Abstract:Traditional smallholder farming systems dominate the savanna range countries of sub-Saharan Africa and provide the foundation for the region's food security. Despite continued expansion of smallholder farming into the surrounding savanna landscapes, food insecurity in the region persists. Central to the monitoring of food security in these countries, and to understanding the processes behind it, are reliable, high-quality datasets of cultivated land. Remote sensing has been frequently used for this purpose but… Show more
“…An example of the combination of different methods for improved classification is given by [55]. The authors combined probabilistic modelling, in the form of logistic regression, with traditional remote sensing approaches to obtain maps of small-scale cropland.…”
Interest in statistical analysis of remote sensing data to produce measurements of environment, agriculture, and sustainable development is established and continues to increase, and this is leading to a growing interaction between the earth science and statistical domains. With this in mind, we reviewed the literature on statistical machine learning methods commonly applied to remote sensing data. We focus particularly on applications related to the United Nations World Bank Sustainable Development Goals, including agriculture (food security), forests (life on land), and water (water quality). We provide a review of useful statistical machine learning methods, how they work in a remote sensing context, and examples of their application to these types of data in the literature. Rather than prescribing particular methods for specific applications, we provide guidance, examples, and case studies from the literature for the remote sensing practitioner and applied statistician. In the supplementary material, we also describe the necessary steps pre and post analysis for remote sensing data; the pre-processing and evaluation steps.
“…An example of the combination of different methods for improved classification is given by [55]. The authors combined probabilistic modelling, in the form of logistic regression, with traditional remote sensing approaches to obtain maps of small-scale cropland.…”
Interest in statistical analysis of remote sensing data to produce measurements of environment, agriculture, and sustainable development is established and continues to increase, and this is leading to a growing interaction between the earth science and statistical domains. With this in mind, we reviewed the literature on statistical machine learning methods commonly applied to remote sensing data. We focus particularly on applications related to the United Nations World Bank Sustainable Development Goals, including agriculture (food security), forests (life on land), and water (water quality). We provide a review of useful statistical machine learning methods, how they work in a remote sensing context, and examples of their application to these types of data in the literature. Rather than prescribing particular methods for specific applications, we provide guidance, examples, and case studies from the literature for the remote sensing practitioner and applied statistician. In the supplementary material, we also describe the necessary steps pre and post analysis for remote sensing data; the pre-processing and evaluation steps.
“…Similarly, accuracies obtained by [ 78 ] were always higher than 80% for sites of intensive farming and stalled at around 50% for sites dominated by smallholder agriculture. While dependable information on commercial farming systems is critical to reduce uncertainty in the global commodity markets, traditional smallholder farming systems dominate the savanna range countries of sub-Saharan Africa and provide the foundation for the regionâs food security [ 79 ]. More generally, estimates suggest that in the rural areas of developing countries around half of the population is smallholder farmers with up to three hectares of cropland [ 80 ].…”
The lack of sufficient ground truth data has always constrained supervised learning, thereby hindering the generation of up-to-date satellite-derived thematic maps. This is all the more true for those applications requiring frequent updates over large areas such as cropland mapping. Therefore, we present a method enabling the automated production of spatially consistent cropland maps at the national scale, based on spectral-temporal features and outdated land cover information. Following an unsupervised approach, this method extracts reliable calibration pixels based on their labels in the outdated map and their spectral signatures. To ensure spatial consistency and coherence in the map, we first propose to generate seamless input images by normalizing the time series and deriving spectral-temporal features that target salient cropland characteristics. Second, we reduce the spatial variability of the class signatures by stratifying the country and by classifying each stratum independently. Finally, we remove speckle with a weighted majority filter accounting for per-pixel classification confidence. Capitalizing on a wall-to-wall validation data set, the method was tested in South Africa using a 16-year old land cover map and multi-sensor Landsat time series. The overall accuracy of the resulting cropland map reached 92%. A spatially explicit validation revealed large variations across the country and suggests that intensive grain-growing areas were better characterized than smallholder farming systems. Informative features in the classification process vary from one stratum to another but features targeting the minimum of vegetation as well as short-wave infrared features were consistently important throughout the country. Overall, the approach showed potential for routinely delivering consistent cropland maps over large areas as required for operational crop monitoring.
“…Table 2 shows that the unrealistic land-use types (which we have classified as "N/A") rarely exist: "paddy on flat sandy land" and "paddy on sand dunes". As mentioned in the introduction, detailed land-use maps are still missing in various regions of the world, including the Horqin Sandy Land, but some remote-sensing researchers have recently worked on the classification of agricultural land uses in some areas, such as Mediterranean and savanna regions [29,[43][44][45][46]. Their research areas and ours share some common land-surface characteristics, e.g., grassland interspersed with trees, shrubs, and crops, which cannot be differentiated by traditional pixel-based classification or vegetation indices [43].…”
Section: Depiction Of Local Land-use Typesmentioning
Previous field research on the Horqin Sandy Land (China), which has suffered from severe desertification during recent decades, revealed how land use on a sand-dune topography affects both land degradation and restoration. This study aimed to depict the spatial distribution of local land use in order to shed more light on previous field findings regarding policies on a broader scale. We performed the following analyses with Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) and Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) images of Advanced Land Observing Satellite (ALOS): (1) object-based classification to discriminate preliminary classification of land-use types that were approximately differentiated by ordinary pixel-based analysis with spectral information; (2) digital photogrammetry to generate a digital surface model (DSM) with adequately high accuracy to represent undulating sand-dune topography; (3) geographic information system (GIS) analysis to classify major topographic types with the digital surface model (DSM); and (4) overlay of the two classification results to depict the local land-use types. The overall accuracies of the object-based and GIS-based classifications were high, at 93% (kappa statistic: 0.84) and 89% (kappa statistic: 0.81), respectively. The resultant local land-use map represents areas covered in previous field studies, showing where and how land degradation and restoration are likely to occur. This research can contribute to future environmental surveys, models, and policies in the study area.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citationsâcitations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.