Integrating Landsat TM and SRTM-DEM derived variables with decision trees for habitat classification and change detection in complex neotropical environments

Remotely sensed images and processing techniques are a primary tool for mapping changes in tropical forest types important to biodiversity and environmental assessment. Detailed land cover data are lacking for most wet tropical areas that present special challenges for data collection. For this study, we utilize decision tree (DT) classifiers to map 32 land cover types of varying ecological and economic importance over an 8000 km² study area and biological corridor in Costa Rica. We assess multivariate QUEST DTs with unbiased classification rules and linear discriminant node models for integrated vegetation mapping and change detection. Predictor variables essential to accurate land cover classification were selected using importance indices statistically derived with classification trees. A set of 35 variables from SRTM-DEM terrain variables, WorldClim grids, and Landsat TM bands were assessed.

Of the techniques examined, QUEST trees were most accurate by integrating a set of 12 spectral and geospatial predictor variables for image subsets with an overall cross-validation accuracy of 93% ± 3.3%. Accuracy with spectral variables alone was low (69% ± 3.3%). A random selection of training and test set pixels for the entire landscape yielded lower classification accuracy (81%) demonstrating a positive effect of image subsets on accuracy. A post-classification change comparison between 1986 and 2001 reveals that two lowland forest types of differing tree species composition are vulnerable to agricultural conversion. Tree plantations and successional vegetation added forest cover over the 15-year time period, but sometimes replaced native forest types, reducing floristic diversity. Decision tree classifiers, capable of combining data from multiple sources, are highly adaptable for mapping and monitoring land cover changes important to biodiversity and other ecosystem services in complex wet tropical environments.     

Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers

Mapping of patterns and spatial distribution of land-use/cover (LULC) has long been based on remotely sensed data. In the recent past, efforts to improve the reliability of LULC maps have seen a proliferation of image classification techniques. Despite these efforts, derived LULC maps are still often judged to be of insufficient quality for operational applications, due to disagreement between generated maps and reference data. In this study we sought to pursue two objectives: first, to test the new-generation multispectral RapidEye imagery classification output using machine-learning random forest (RF) and support vector machines (SVM) classifiers in a heterogeneous coastal landscape; and second, to determine the importance of different RapidEye bands on classification output. Accuracy of the derived thematic maps was assessed by computing confusion matrices of the classifiers’ cover maps with respective independent validation data sets. An overall classification accuracy of 93.07% with a kappa value of 0.92, and 91.80 with a kappa value of 0.92 was achieved using RF and SVM, respectively. In this study, RF and SVM classifiers performed comparatively similarly as demonstrated by the results of McNemer’s test (Z = 1.15). An evaluation of different RapidEye bands using the two classifiers showed that incorporation of the red-edge band has a significant effect on the overall classification accuracy in vegetation cover types. Consequently, pursuit of high classification accuracy using high-spatial resolution imagery on complex landscapes remains paramount.     

A survey of methods incorporating spatial information in image classification and spectral unmixing

Over the past decade, the incorporation of spatial information has drawn increasing attention in multispectral and hyperspectral data analysis. In particular, the property of spatial autocorrelation among pixels has shown great potential for improving understanding of remotely sensed imagery. In this paper, we provide a comprehensive review of the state-of-the-art techniques in incorporating spatial information in image classification and spectral unmixing. For image classification, spatial information is accounted for in the stages of pre-classification, sample selection, classifiers, post-classification, and accuracy assessment. With regards to spectral unmixing, spatial information is discussed in the context of endmember extraction, selection of endmember combinations, and abundance estimation. Finally, a perspective on future research directions for advancing spatial-spectral methods is offered.     

Pre-classification and post-classification change-detection techniques to monitor land-cover and land-use change using multi-temporal Landsat imagery: a case study on Pisa Province in Italy

This article explores the simultaneous use of pre-classification and post-classification change-detection techniques to map and monitor land-cover and land-use change using multi-temporal Landsat Multi-spectral Scanner and Enhanced Thematic Mapper plus data over one of the most important tourism centres of Italy (e.g. Pisa Province) for 1972, 2000 and 2006.

Pre-classification approaches of principal-component analysis and band combination are potentially tailored to reduce data redundancy of the satellite imagery in order to highlight different objects of significance for change-detection analysis across time-series data. In this work, the application of pre-classification techniques could contribute to produce land-cover and land-use maps with higher quality of classification. At this point, the average value of overall classification accuracies for the three classification outputs was an estimated 90%.

Then, ‘from–to’ change information, as well as the area and the type of landscape transformations, are provided through the post-classification technique. The findings of this study show that the province of Pisa has significantly experienced a high rate of deforestation and urban development over past decades. It is revealed that artificial structures (e.g. urban and industrial zones) in Pisa Province increased at a change rate of around 265% and forested land decreased from approximately 45% to 32% of the total area of the province between 1972 and 2006. Likewise, the perceptible growth of built-up structures from about 4% to 10.6% in Pisa City during this 34-year period has imposed a heavy pressure on the landscape of Pisa.     

Guidance on and comparison of machine learning classifiers for Landsat-based land cover and land use mapping

Remote sensing scientists are increasingly adopting machine learning classifiers for land cover and land use (LCLU) mapping, but model selection, a critical step of the machine learning classification, has usually been ignored in the past research. In this paper, step-by-step guidance (for classifier training, model selection, and map production) with supervised learning model selection is first provided. Then, model selection is exhaustively applied to different machine learning (e.g. Artificial Neural Network (ANN), Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF)) classifiers to identify optimal polynomial degree of input features (d) and hyperparameters with Landsat imagery of a study region in China and Ghana. We evaluated the map accuracy and computing time associated with different versions of machine learning classification software (i.e. ArcMap, ENVI, TerrSet, and R).

The optimal classifiers and their associated polynomial degree of input features and hyperparameters vary for the two image datasets that were tested. The optimum combination of d and hyperparameters for each type of classifier was used across software packages, but some classifiers (i.e. ENVI and TerrSet ANN) were customized due to the constraints of software packages. The LCLU map derived from ENVI SVM has the highest overall accuracy (72.6%) for the Ghana dataset, while the LCLU map derived from R DT has the highest overall accuracy (48.0%) for the FNNR dataset. All LCLU maps for the Ghana dataset are more accurate compared to those from the China dataset, likely due to more limited and uncertain training data for the China (FNNR) dataset. For the Ghana dataset, LCLU maps derived from tree-based classifiers (ArcMap RF, TerrSet DT, and R RF) routines are accurate, but these maps have artefacts resulting from model overfitting problems.     

A hybrid approach to urban land use/cover mapping using Landsat 7 Enhanced Thematic Mapper Plus (ETM+) images

A hybrid method that incorporates the advantages of supervised and unsupervised approaches as well as hard and soft classifications was proposed for mapping the land use/cover of the Atlanta metropolitan area using Landsat 7 Enhanced Thematic Mapper Plus (ETM+) data. The unsupervised ISODATA clustering method was initially used to segment the image into a large number of clusters of pixels. With reference to ground data based on 1?:?40?000 colour infrared aerial photographs in the form of Digital Orthophoto Quarter Quadrangle (DOQQ), homogeneous clusters were labelled. Clusters that could not be labelled because of mixed pixels were clipped out and subjected to a supervised fuzzy classification. A final land use/cover map was obtained by a union overlay of the two partial land use/cover maps. This map was evaluated by comparing with maps produced using unsupervised ISODATA clustering, supervised fuzzy and supervised maximum likelihood classification methods. It was found that the hybrid approach was slightly better than the unsupervised ISODATA clustering in land use/cover classification accuracy, most probably because of the supervised fuzzy classification, which effectively dealt with the mixed pixel problem in the low-density urban use category of land use/cover. It was suggested that this hybrid approach can be economically implemented in a standard image processing software package to produce land use/cover maps with higher accuracy from satellite images of moderate spatial resolution in a complex urban environment, where both discrete and continuous land cover elements occur side by side.     

Comparison of pixel- and object-based classification in land cover change mapping

Land use/land cover (LULC) change occurs when humans alter the landscape, and this leads to increasing loss, fragmentation and spatial simplification of habitat. Many fields of study require monitoring of LULC change at a variety of scales. LULC change assessment is dependent upon high-quality input data, most often from remote sensing-derived products such as thematic maps. This research compares pixel- and object-based classifications of Landsat Thematic Mapper (TM) data for mapping and analysis of LULC change in the mixed land use region of eastern Ontario for the period 1995–2005. For single date thematic maps of 10 LULC classes, quantitative and visual analyses showed no significant accuracy difference between the two methods. The object-based method produced thematic maps with more uniform and meaningful LULC objects, but it suffered from absorption of small rare classes into larger objects and the incapability of spatial parameters (e.g. object shape) to contribute to class discrimination. Despite the similar map accuracies produced by the two methods, temporal change maps produced using post-classification comparison (PCC) and analysed using intensive visual analysis of errors of omission and commission revealed that the object-based maps depicted change more accurately than maximum likelihood classification (MLC)-derived change maps.     

Optimizing unsupervised classifications of remotely sensed imagery with a data-assisted labeling approach

The quality of remotely sensed land use and land cover (LULC) maps is affected by the accuracy of image data classifications. Various efforts have been made in advancing supervised or unsupervised classification methods to increase the repeatability and accuracy of LULC mapping. This study incorporates a data-assisted labeling approach (DALA) into the unsupervised classification of remotely sensed imagery. The DALA-unsupervised classification algorithm consists of three steps: (1) creation of N spectral-class maps using Iterative Self-Organizing Data Analysis Technique Algorithm (ISODATA); (2) development of LULC maps with assistance of reference data; and (3) accuracy assessments of all the LULC maps using independent reference data and selection of one LULC map with the highest accuracy. Classification experiments with a composite image of a Landsat Thematic Mapper (TM) image and an Enhanced Thematic Mapper Plus (ETM+) image suggest that DALA was effective in making unsupervised classification process more objective, automatic, and accurate. A comparison between the DALA-unsupervised classifications and some conventional classifications suggests that the DALA-unsupervised classification algorithm yielded better classification accuracies compared to these conventional approaches. Such a simple, effective approach has not been systematically examined before but has great potential for many applications in the geosciences.     

Comparison of fuzzy c-means classification,linear mixture modelling and MLC probabilities as tools for unmixing coarse pixels

Three different 'soft' classifiers (fuzzy c-means classifier, linear mixture model, and probability values from a maximum likelihood classification) were used for unmixing of coarse pixel signatures to identify four land cover classes (i.e., supervised classifications). The coarse images were generated from a 30m Thematic Mapper (TM) image; one set by mean filtering, and another using an asymmetric filter kernel to simulate Multi-Spectral Scanner (MSS) sensor sampling. These filters collapsed together windows of up to 11 11 pixels. The fractional maps generated by the three classifiers were compared to truth maps at the corresponding scales, and to the results of a hard maximum likelihood classification. Overall, the fuzzy c-means classifier gave the best predictions of sub-pixel landcover areas, followed by the linear mixture model. The probabilities differed little from the hard classification, suggesting that the clusters should be modelled more loosely. This paper demonstrates successful methods for use and comparison of the classifiers that should ideally be extended to a real dataset.     

基于C5.0决策树算法的西北干旱区土地覆盖分类研究——以甘肃省武威市为例

西北干旱区面积广阔,由于土地利用类型多样,成因复杂,对环境变化敏感、变化过程快、幅度大、景观差异明显等特点,在影像上表现出的“同物异谱”现象明显 |利用常规目视解译、监督非监督分类、人工参与的决策树分类等方法在效率或精度等方面各有其缺陷。采用机器学习C5.0决策树算法,综合利用地物波谱、NDVI、TC、纹理等信息,根据样本数据自动挖掘分类规则并对整个研究区进行地物分类。机器学习的决策树可以挖掘出更多的分类规则,C5.0算法对采样数据的分布没有要求,可以处理离散和连续数据,生成的规则易于理解,分类精度高,可以满足西北干旱区大面积的土地利用/覆被变化制图的需要。     

SVM在多源遥感图像分类中的应用研究

在利用遥感图像进行土地利用/覆盖分类过程中,可采用以下两种途径来提高分类精度:一是通过增加有利于分类的数据源,引入地理辅助数据和归一化植被指数(NDVI)来进行多源信息融合;二是选择更好的分类方法,例如支持向量机(SVM)学习方法,由于该方法克服了最大似然法和神经网络的弱点,非常适合高维、复杂的小样本多源数据的分类。为了提高多源遥感图像分类的精度,还研究了支持向量机在遥感图像分类中模型的选择,包括多类模型和核函数的选择。分类结果表明,支持向量机比传统的分类方法具有更高的精度,尤其是基于径向基核函数和一对一多类方法的支持向量机模型更适合多源遥感图像分类,因此,基于支持向量机的多源土地利用/覆盖分类能大大提高分类精度。     

The IGBP-DIS global 1km land cover data set,DISCover: First results

The International Geosphere-Biosphere Programme Data and Information System (IGBP-DIS) is co-ordinating the development of global land data sets from Advanced Very High Resolution Radiometer (AVHRR) data. The first is a 1 km spatial resolution land cover product 'DISCover', based on monthly Normalized Difference Vegetation Index composites from 1992 and 1993. DISCover is a 17 class land cover dataset based on the science requirements of IGBP elements. Mapping uses unsupervised classification with post-classification refinement using ancillary data. Draft Africa, North America and South America products are now available for peer review.     

Supervised image classification by MLP and RBF neural networks with and without an exhaustively defined set of classes

The absence of assumptions about the dataset to be classified is one of the major attractions of neural networks for supervised image classification applications. Classification by a neural network does, however, make assumptions about the classes. One key assumption typically made is that the set of classes has been defined exhaustively. If this assumption is unsatisfied, cases of an untrained class will be present and commissioned into the set of trained classes to the detriment of classification accuracy. This was observed in land cover classifications derived with multi-layer perceptron (MLP) and radial basis function (RBF) neural networks in which the presence of an untrained class resulted in a ～12.5% decrease in the accuracy of crop classifications derived from airborne thematic mapper data. However, since the RBF network partitions feature space locally rather than globally as with the MLP, it was possible to reduce the commission of atypical cases into the set of trained classes through the setting of post-classification thresholds on the RBF network's outputs. As a result it was possible to identify and exclude some cases of untrained classes from a classification with a RBF network which resulted in an increase in classification accuracy.     

基于多分类器集成的“北京一号”小卫星遥感影像分类研究

将多分类器集合应用于"北京一号"小卫星多光谱遥感数据土地覆盖分类,首先构建分类器集合,应用最小距离分类、最大似然分类、支持向量机(SVM)、BP神经网络、RBF神经网络和决策树等进行土地覆盖分类,然后利用Bagging、Boosting、投票法、证据理论和模糊积分法等分类器集成方法,得到综合不同分类器输出的最终分类结果。试验表明,多分类器集成能够有效提高"北京一号"小卫星土地覆盖分类的精度,具有广泛的应用前景。     

Approaches for the production and evaluation of fuzzy land cover classifications from remotely-sensed data

Remote sensing is an attractive source of data for land cover mapping applications. Mapping is generally achieved through the application of a conventional statistical classification, which allocates each image pixel to a land cover class. Such approaches are inappropriate for mixed pixels, which contain two or more land cover classes, and a fuzzy classification approach is required. When pixels may have multiple and partial class membership measures of the strength of class membership may be output and, if strongly related to the land cover composition, mapped to represent such fuzzy land cover. This type of representation can be derived by softening the output of a conventional ‘hard’ classification or using a fuzzy classification. The accuracy of the representation provided by a fuzzy classification is, however, difficult to evaluate. Conventional measures of classification accuracy cannot be used as they are appropriate only for ‘hard’ classifications. The accuracy of a classification may, however, be indicated by the way in which the strength of class membership is partitioned between the classes and how closely this represents the partitioning of class membership on the ground. In this paper two measures of the closeness of the land cover representation derived from a classification to that on the ground were used to evaluate a set of fuzzy classifications. The latter were based on measures of the strength of class membership output from classifications by a discriminant analysis, artificial neural network and fuzzy c-means classifiers. The results show the importance of recognising and accommodating for the fuzziness of the land cover on the ground. The accuracy assessment methods used were applicable to pure and mixed pixels and enabled the identification of the most accurate land cover representation derived. The results showed that the fuzzy representations were more accurate than the ‘hard’ classifications. Moreover, the outputs derived from the artificial neural network and the fuzzy c-means algorithm in particular were strongly related to the land cover on the ground and provided the most accurate land cover representations. The ability to appropriately represent fuzzy land cover and evaluate the accuracy of the representation should facilitate the use of remote sensing as a source of land cover data.     

Detection of land use/land cover changes for the northern part of the Nile delta (Burullus region), Egypt

The monitoring of land use/land cover changes along the north part of the Nile delta, Egypt is very important for the planner, management, governmental and non-governmental organizations and the scientific community. This information is essential for planning and implementing policies to optimize the use of natural resources and accommodate development whilst minimizing the impact on the environment. To study these changes along the study area, two sets of Landsat Thematic Mapper (TM) data were used. TM data are useful for this type of study due to its high spatial resolution, spectral resolution and low repetitive acquisition (16 days). A post-classification technique is used in this study based on hybrid classification (unsupervised and supervised). Each method used was assessed, and checked in field. Nine land use/land cover classes are produced. The overall accuracy for a 1984 image is 78% and for a 1997 image is 80%. The objective of this study was to detect land use/land cover changes, and to assess the nature of future change following construction of the international coastal road which crosses the study area.     

An evaluation of fuzzy and texture-based classification approaches for mapping regenerating tropical forest classes from Landsat-TM data

Two classification approaches were investigated for the mapping of tropical forests from Landsat-TM data of a region north of Manaus in the Brazilian state of Amazonas. These incorporated textural information and made use of fuzzy approaches to classification. In eleven class classifications the texture-based classifiers (based on a Markov random field model) consistently provided higher classification accuracies than conventional per-pixel maximum likelihood and minimum distance classifications, indicating that they are more able to characterize accurately several regenerating forest classes. Measures of the strength of class memberships derived from three classification algorithms (based on the probability density function, a posteriori probability and the Mahalanobis distance) could be used to derive fuzzy image classifications and be used in post-classification processing. The latter, involving either the summation of class memberships over a local neighbourhood or the application of homogeneity measures, were found to increase classification accuracy by some 10 per cent in comparison with a conventional maximum likelihood classification, a result of comparable accuracy to that derived from the texture-based classifications.     

Mapping Land Cover from Remotely Sensed Data with a Softened Feedforward Neural Network Classification

Remote sensing has considerable potential as a source of data for land cover mapping. This potential remains to be fully realised due, in part, to the methods used to extract land cover information from the remotely sensed data. Widely used statistical classifiers provide a poor representation of land cover, make untenable assumptions about the data and convey no information on the quality of individual class allocations. This paper shows that a softened classification, providing information on the strength of membership to all classes for each image pixel, may be derived from a neural network. This information may be used to indicate classification quality on a per-pixel basis. Moreover, a soft or fuzzy classification may be derived to more appropriately represent land cover than the conventional hard classification.     

Evaluation of MK-4 multispectral satellite photography in land cover classification of eastern Ontario

Russian MK-4 multispectral satellite photography has been investigated for potential in land cover classification. Thematic maps were generated using maximum likelihood, neural network and context classifiers. Classifications of the raw spectral data, of spectral transforms, and of combined spectral/textural data were evaluated. Low point-based class accuracies resulted for land cover types exhibiting high spatial variability at the given pixel spacing of 7.5m, while more spatially homogeneous cover types were well classified. Several issues arose which need to be addressed for effective future use of high-resolution satellite sensors in regional land cover mapping. They include the need for further research in techniques for classification and accuracy assessment which are sensitive to the spatial variance of such high resolution imagery, and optimization of class attribute definitions.     

Feature selection and land cover classification of a MODIS-like data set for a semiarid environment

The advent of the Earth Observing System (EOS), and the Moderateresolution Imaging Spectroradiometer (MODIS) in particular, will usher in a new era of global remote sensing by providing very large data volumes for interpretation and processing. Since many data streams will contain correlated data, feature selection is an important practical problem for such activities as classification of global land cover based on spectral, temporal, spatial and directional data. Treebased classification methods offer a suite of promising approaches to extraction of meaningful features from large measurement spaces. This research develops a tree-based model that performs feature selection on a satellite database containing information on land covers in a semiarid region in Cochise County, Arizona. In addition, we test the abilities of several classifiers to correctly label land cover using this reduced set of inputs under various sampling schemes. Results from this analysis indicate that decision trees can reduce a high-dimension dataset to a manageable set of inputs that retain most of the information of the original database, while remaining largely insensitive to choice of sampling strategy, and that Fuzzy ARTMAP, a type of artificial neural network classifier, achieves highest accuracy in comparison to maximum-likelihood or decision-tree classifiers.