Conference Paper

A Maximum Entropy Approach for Mapping Falcata Plantations in Sentinel-2 Imagery

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Multiple-class land-cover classification approaches can be inefficient when the main goal is to classify only one or a few classes. Under this scenario one-class classification algorithms could be a more efficient alternative. Currently there are several algorithms that can fulfil this task, with MaxEnt being one of the most promising. However, there is scarce information regarding parametrization for performing land-cover classification using MaxEnt. In this study we aimed to understand how MaxEnt parameterization affects the classification accuracy of four different land-covers (i.e., built-up, irrigated grass, evergreen trees and deciduous trees) in the city of Santiago de Chile. We also evaluated if MaxEnt manual parameterization outperforms classification results obtained when using MaxEnt default parameters setting. To accomplish our objectives, we generated a set of 25,344 classification maps (i.e., 6,336 for each assessed land-cover), which are based on all the potential combination of 12 different classes of features restrictions, four regularization multipliers, four different sample sizes, three training/testing proportions, and 11 thresholds for generating the binary maps. Our results showed that with a good parameterization, MaxEnt can effectively classify different land covers with kappa values ranging from 0.68 for deciduous trees to 0.89 for irrigated grass. However, the accuracy of classification results is highly influenced by the type of land-cover being classified. Simpler models produced good classification outcomes for homogenous land-covers, but not for heterogeneous covers, where complex models provided better outcomes. In general, manual parameterization improves the accuracy of classification results, but this improvement will depend on the threshold used to generate the binary map. In fact, threshold selection showed to be the most relevant factor impacting the accuracy of the four land-cover classification. The number of sampling points for training the model also has a positive effect on classification results. However, this effect followed a logarithmic distribution, showing an improvement of kappa values when increasing the sampling from 40 to 60 points, but showing only a marginal effect if more than 60 sampling points are used. In light of these results, we suggest testing different parametrization and thresholds until satisfactory kappa or other accuracy metrics values are achieved. Our results highlight the huge potential that MaxEnt has a as a tool for one-class classification, but a good understanding of the software settings and model parameterization is needed to obtain reliable results.
Article
Full-text available
Aboveground biomass estimation (AGB) is essential in determining the environmental and economic values of mangrove forests. Biomass prediction models can be developed through integration of remote sensing, field data and statistical models. This study aims to assess and compare the biomass predictor potential of multispectral bands, vegetation indices and biophysical variables that can be derived from three optical satellite systems: the Sentinel-2 with 10 m, 20 m and 60 m resolution; RapidEye with 5m resolution and PlanetScope with 3m ground resolution. Field data for biomass were collected from a Rhizophoraceae-dominated mangrove forest in Masinloc, Zambales, Philippines where 30 test plots (1.2 ha) and 5 validation plots (0.2 ha) were established. Prior to the generation of indices, images from the three satellite systems were pre-processed using atmospheric correction tools in SNAP (Sentinel-2), ENVI (RapidEye) and python (PlanetScope). The major predictor bands tested are Blue, Green and Red, which are present in the three systems; and Red-edge band from Sentinel-2 and Rapideye. The tested vegetation index predictors are Normalized Differenced Vegetation Index (NDVI), Soil-adjusted Vegetation Index (SAVI), Green-NDVI (GNDVI), Simple Ratio (SR), and Red-edge Simple Ratio (SRre). The study generated prediction models through conventional linear regression and multivariate regression. Higher coefficient of determination (r²) values were obtained using multispectral band predictors for Sentinel-2 (r² = 0.89) and Planetscope (r² = 0.80); and vegetation indices for RapidEye (r² = 0.92). Multivariate Adaptive Regression Spline (MARS) models performed better than the linear regression models with r² ranging from 0.62 to 0.92. Based on the r² and root-mean-square errors (RMSE’s), the best biomass prediction model per satellite were chosen and maps were generated. The accuracy of predicted biomass maps were high for both Sentinel-2 (r² = 0.92) and RapidEye data (r² = 0.91).
Article
Full-text available
Spartina alterniflora has become the main invasive plant along the Chinese coast and now threatens the local ecological environment. Accurately monitoring the distribution of S. alterniflora is urgent and essential for developing cost-effective control strategies. In this study, we applied the One-Class Classification (OCC) methods of Maximum entropy (Maxent) and Biased Support Vector Machine (BSVM) based on Landsat time-series imagery to detect the species on the middle coast of Jiangsu in east China. We conducted four experimental setups (i.e., single-scene analysis, time-series analysis, Normalized Difference Vegetation Index (NDVI) time-series analysis and a compressed time-series analysis), using OCC methods to recognize the species. Then, we tested the performance of a compressed time-series model for S. alterniflora detection and evaluated the expansibility of this approach when it was applied to a larger region. Our principal findings are as follows: (1) Maxent and BSVM performed equally well, and Maxent appeared to have a more balanced performance over the summer months; (2) the Maxent model with the Default Parameter Set (Maxent-DPS) showed a slightly higher accuracy and more overfitting than Maxent with the Akaike Information Criterion corrected for small samples sizes (AICc)-selected parameter set model, but a t-test found no significant difference between these two settings; (3) April and December were deemed to be important periods for the detection of S. alterniflora; (4) a compressed time-series analysis model—including only three variables (December NDVI, March green and the third Principal Component in January, PC3)—yielded higher accuracy than single-scene analyses, which indicated that time-series analysis can better detect S. alterniflora than single-scene analyses; and (5) the Maxent model using the reconstructed optimal variables and 70 training samples over a larger region produced encouraging results with an overall accuracy of 90.88% and a Kappa of 0.78. The one-class classification method combined with a phenology-based detection strategy is therefore promising for the application of the long-term detection of S. alterniflora over extended areas.
Article
Full-text available
In this study we used the maximum entropy (MAXENT) method to test the extent to which a combination of temporal NDVI series and ancillary field data can be used to map the distribution of maize fields, soybean fields, and tobacco fields. We also test whether there are specific times during the growing season suitable for separately mapping the fields. Results indicate that we can separately map fields of maize, soybeans, and tobacco using remotely sensed data at specific times during the growing season. We conclude that MAXENT can be used to map fields of maize, soybeans, and tobacco using a temporal series of NDVI data. In addition, we identify critical periods that provide significant (p < 0.05) NDVI values useful for mapping.
Article
Full-text available
Terracing is a basic component of agricultural land use systems across multiple landscapes and time periods. This study develops a predictive model using the environmental variables that differentiate ancient and modern terrace locations in the foothills of the Troodos Mountains, Cyprus. Model development utilized the maximum entropy principle as applied by Maxent software to estimate probability distributions for both terrace types across geographical space based upon presence-only observations and environmental variables. The Maxent models are effective in predicting potential terrace distributions and reveal systematic differences between ancient and modern terrace arrays in the study area. Surface geology played a key role in predicting ancient terrace locations, but contributed much less substantially in modeling modern terraces, whose locations were influenced strongly by topography. Maxent modeling is shown to be effective in assessing environmental constraints and terrace locations, providing a method that can be used by physical geographers and archaeologists for modeling human–environment interactions in agrarian societies.
Article
Full-text available
In this study, we tested the Maximum Entropy model (Maxent) for its application and performance in remotely sensing invasive Tamarix sp. Six Landsat 7 ETM+ satellite scenes and a suite of vegetation indices at different times of the growing season were selected for our study area along the Arkansas River in Colorado. Satellite scenes were selected for April, May, June, August, September, and October and tested in single-scene and time-series analyses. The best model was a time-series analysis fit with all spectral variables, which had an AUC = 0.96, overall accuracy = 0.90, and Kappa = 0.79. The top predictor variables were June tasselled cap wetness, September tasselled cap wetness, and October band 3. A second time-series analysis, where the variables that were highly correlated and demonstrated low predictive strengths were removed, was the second best model. The third best model was the October single-scene analysis. Our results may prove to be an effective approach for mapping Tamarix sp., which has been a challenge for resource managers. Of equal importance is the positive performance of the Maxent model in handling remotely sensed datasets.
Article
The MaxEnt software package is one of the most popular tools for species distribution and environmental niche modeling, with over 1000 published applications since 2006. Its popularity is likely for two reasons: 1) MaxEnt typically outperforms other methods based on predictive accuracy and 2) the software is particularly easy to use. MaxEnt users must make a number of decisions about how they should select their input data and choose from a wide variety of settings in the software package to build models from these data. The underlying basis for making these decisions is unclear in many studies, and default settings are apparently chosen, even though alternative settings are often more appropriate. In this paper, we provide a detailed explanation of how MaxEnt works and a prospectus on modeling options to enable users to make informed decisions when preparing data, choosing settings and interpreting output. We explain how the choice of background samples reflects prior assumptions, how nonlinear functions of environmental variables (features) are created and selected, how to account for environmentally biased sampling, the interpretation of the various types of model output and the challenges for model evaluation. We demonstrate MaxEnt’s calculations using both simplified simulated data and occurrence data from South Africa on species of the flowering plant family Proteaceae. Throughout, we show how MaxEnt’s outputs vary in response to different settings to highlight the need for making biologically motivated modeling decisions.
Article
Mapping of habitats with relevance for nature conservation involves the identification of patches of target habitats in a complex mosaic of vegetation types not relevant for conservation planning. Limiting the necessary ground reference to a small sample of target habitats would greatly reduce and therefore support the field mapping effort. We thus aim to answer in this study the question: can semi-automated remote sensing methods help to map such patches without the need of ground references from sites not relevant for nature conservation? Approaches able to fulfill this task may help to improve the efficiency of large scale mapping and monitoring programs such as requested for the European Habitat Directive. In the present study, we used the maximum-entropy based classification approach Maxent to map four habitat types across a patchy landscape of 1000 km2 near Munich, Germany. This task was conducted using the low number of 125 ground reference points only along with easily available multi-seasonal RapidEye satellite imagery. Encountered problems include the non-stationarity of habitat reflectance due to different phenological development across space, continuous transitions between the habitats and the need for improved methods for detailed validation. The result of the tested approach is a habitat map with an overall accuracy of 70%. The rather simple and affordable approach can thus be recommended for a first survey of previously unmapped areas, as a tool for identifying potential gaps in existing habitat inventories and as a first check for changes in the distribution of habitats.
Article
In remote sensing classification there are situations when users are only interested in classifying one specific land type without considering other classes, which is referred to as one-class classification. Traditional supervised learning requires all classes that occur in the image to be exhaustively labelled and hence is inefficient for one-class classification. In this study we investigate a maximum entropy approach (MAXENT) to one-class classification of remote sensing imagery, i.e. classifying a single land class (e.g. urban areas, trees, grasses and soils) from an aerial photograph with 0.3 m spatial resolution. MAXENT estimates the Gibbs probability distribution that is proportional to the conditional probability of being positive. A threshold for generating binary predictions can be determined based on the omission rate of a validation set. The results indicate that MAXENT provides higher classification accuracy than the one-class support vector machine (OCSVM). MAXENT does not require other land classes for training. Its input is only a set of training samples of the specific land class of interest, as well as a set of known constraints on the distribution. Therefore, the effort of manually collecting training data for classification can be significantly reduced.
Article
The availability of detailed environmental data, together with inexpensive and powerful computers, has fueled a rapid increase in predictive modeling of species environmental requirements and geographic distributions. For some species, detailed presence/absence occurrence data are available, allowing the use of a variety of standard statistical techniques. However, absence data are not available for most species. In this paper, we introduce the use of the maximum entropy method (Maxent) for modeling species geographic distributions with presence-only data. Maxent is a general-purpose machine learning method with a simple and precise mathematical formulation, and it has a number of aspects that make it well-suited for species distribution modeling. In order to investigate the efficacy of the method, here we perform a continental-scale case study using two Neotropical mammals: a lowland species of sloth, Bradypus variegatus, and a small montane murid rodent, Microryzomys minutus. We compared Maxent predictions with those of a commonly used presence-only modeling method, the Genetic Algorithm for Rule-Set Prediction (GARP). We made predictions on 10 random subsets of the occurrence records for both species, and then used the remaining localities for testing. Both algorithms provided reasonable estimates of the species’ range, far superior to the shaded outline maps available in field guides. All models were significantly better than random in both binomial tests of omission and receiver operating characteristic (ROC) analyses. The area under the ROC curve (AUC) was almost always higher for Maxent, indicating better discrimination of suitable versus unsuitable areas for the species. The Maxent modeling approach can be used in its present form for many applications with presence-only datasets, and merits further research and development.
Article
Remote sensing is a major source of land-cover information. Commonly, interest focuses on a single land-cover class. Although a conventional multiclass classifier may be used to provide a map depicting the class of interest, the analysis is not focused on that class and may be suboptimal in terms of the accuracy of its classification. With a conventional classifier, considerable effort is directed on the classes that are not of interest. Here, it is suggested that a one-class-classification approach could be appropriate when interest focuses on a specific class. This is illustrated with the classification of fenland, a habitat of considerable conservation value, from Landsat Enhanced Thematic Mapper Plus imagery. A range of one-class classifiers is evaluated, but attention focuses on the support-vector data description (SVDD). The SVDD was used to classify fenland with an accuracy of 97.5% and 93.6% from the user's and producer's perspectives, respectively. This classification was trained upon only the fenland class and was substantially more accurate in fen classification than a conventional multiclass maximum-likelihood classification provided with the same amount of training data, which classified fen with an accuracy of 90.0% and 72.0% from the user's and producer's perspectives, respectively. The results highlight the ability to classify a single class using only training data for that class. With a one-class classification, the analysis focuses tightly on the class of interest, with resources and effort not directed on other classes, and there are opportunities to derive highly accurate classifications from small training sets
Evaluation of One-Class Classification Methods for Mapping Rubber Plantation in Agusan del Sur
  • J T Versoza
The Philippine’s Wood Industry
  • E G C Zaragoza
Maximum entropy modeling of species geographic distributions
  • S J Phillips
  • R P Anderson
  • R E Schapire
The Philippine&#x2019;s Wood Industry
  • zaragoza