Development of a modified neural network-based land cover classification system using automated data selector and multiresolution remotely sensed data

Integrating multiple images with artificial neural networks (ANN) improves classification accuracy. ANN performance is sensitive to training datasets. Complexity and errors compound when merging multiple data, pointing to needs for new techniques. Kohonen's self-organizing mapping (KSOM) neural network was adapted as an automated data selector (ADS) to replace manual training data processes. The multilayer perceptron (MLP) network was then trained using automatically extracted datasets and used for classification. Two hypotheses were tested: ADS adapted from the KSOM network provides adequate and reliable training datasets, improving MLP classification performance; and fusion of Landsat thematic mapper (TM) and SPOT images using the modified ANN approach increases accuracy. ADS adapted from the KSOM network improved training data quality and increased classification accuracy and efficiency. Fusion of compatible multiple data can improve performance if appropriate training datasets are collected. This proved to be a viable classification scheme particularly where acquiring sufficient and reliable training datasets is difficult.     

Spatial application of Random Forest models for fine-scale coastal vegetation classification using object based analysis of aerial orthophoto and DEM data

High spatial resolution mapping of natural resources is much needed for monitoring and management of species, habitats and landscapes. Generally, detailed surveillance has been conducted as fieldwork, numerical analysis of satellite images or manual interpretation of aerial images, but methods of object-based image analysis (OBIA) and machine learning have recently produced promising examples of automated classifications of aerial imagery. The spatial application potential of such models is however still questionable since the transferability has rarely been evaluated.We investigated the potential of mosaic aerial orthophoto red, green and blue (RGB)/near infrared (NIR) imagery and digital elevation model (DEM) data for mapping very fine-scale vegetation structure in semi-natural terrestrial coastal areas in Denmark. The Random Forest (RF) algorithm, with a wide range of object-derived image and DEM variables, was applied for classification of vegetation structure types using two hierarchical levels of complexity. Models were constructed and validated by cross-validation using three scenarios: (1) training and validation data without spatial separation, (2) training and validation data spatially separated within sites, and (3) training and validation data spatially separated between different sites.Without spatial separation of training and validation data, high classification accuracies of coastal structures of 92.1% and 91.8% were achieved on coarse and fine thematic levels, respectively. When models were applied to spatially separated observations within sites classification accuracies dropped to 85.8% accuracy at the coarse thematic level, and 81.9% at the fine thematic level. When the models were applied to observations from other sites than those trained upon the ability to discriminate vegetation structures was low, with 69.0% and 54.2% accuracy at the coarse and fine thematic levels, respectively.Evaluating classification models with different degrees of spatial correlation between training and validation data was shown to give highly different prediction accuracies, thereby highlighting model transferability and application potential. Aerial image and DEM-based RF models had low transferability to new areas due to lack of representation of aerial image, landscape and vegetation variation in training data. They do, however, show promise at local scale for supporting conservation and management with vegetation mappings of high spatial and thematic detail based on low-cost image data.     

Remotely sensed characterization of forest fuel types by using satellite ASTER data

The characterization of fuel types is very important for computing spatial fire hazard and risk and simulating fire growth and intensity across a landscape. However, due to the complex nature of fuel characteristic a fuel map is considered one of the most difficult thematic layers to build up. The advent of sensors with increased spatial resolution may improve the accuracy and reduce the cost of fuels mapping. The objective of this research is to evaluate the accuracy and utility of imagery from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite imagery. In order to ascertain how well ASTER data can provide an exhaustive classification of fuel properties a sample area characterized by mixed vegetation covers was analysed. The selected sample areas has an extension at around 60 km² and is located inside the Sila plateau in the Calabria Region (South of Italy). Fieldwork fuel type recognitions, performed before, after and during the acquisition of remote sensing ASTER data, were used as ground-truth dataset to assess the results obtained for the considered test area. The method comprised the following three steps: (I) adaptation of Prometheus fuel types for obtaining a standardization system useful for remotely sensed classification of fuel types and properties in the considered Mediterranean ecosystems; (II) model construction for the spectral characterization and mapping of fuel types based on a maximum likelihood (ML) classification algorithm; (III) accuracy assessment for the performance evaluation based on the comparison of ASTER-based results with ground-truth. Results from our analysis showed that the use ASTER data provided a valuable characterization and mapping of fuel types being that the achieved classification accuracy was higher than 90%.     

Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using Support Vector Machines

Crop mapping is one major component of agricultural resource monitoring using remote sensing. Yield or water demand modeling requires that both, the total surface that is cultivated and the accurate distribution of crops, respectively is known. Map quality is crucial and influences the model outputs. Although the use of multi-spectral time series data in crop mapping has been acknowledged, the potentially high dimensionality of the input data remains an issue. In this study Support Vector Machines (SVM) are used for crop classification in irrigated landscapes at the object-level. Input to the classifications is 71 multi-seasonal spectral and geostatistical features computed from RapidEye time series. The random forest (RF) feature importance score was used to select a subset of features that achieved optimal accuracies. The relationship between the hard result accuracy and the soft output from the SVM is investigated by employing two measures of uncertainty, the maximum a posteriori probability and the alpha quadratic entropy. Specifically the effect of feature selection on map uncertainty is investigated by looking at the soft outputs of the SVM, in addition to classical accuracy metrics. Overall the SVMs applied to the reduced feature subspaces that were composed of the most informative multi-seasonal features led to a clear increase in classification accuracy up to 4.3%, and to a significant decline in thematic uncertainty. SVM was shown to be affected by feature space size and could benefit from RF-based feature selection. Uncertainty measures from SVM are an informative source of information on the spatial distribution of error in the crop maps.     

基于层次性邻近分析的农用地分等数据库综合

以农用地分等数据库综合为例,提出了图斑归并同时顾及几何距离和多种专题属性的层次性邻近分析方法,及基于层次性邻近分析的多边形聚合和融合方法,并以实例验证了其有效性.     

Measures of spatio-temporal accuracy for time series land cover data

Remote sensing is a useful tool for monitoring changes in land cover over time. The accuracy of such time-series analyses has hitherto only been assessed using confusion matrices. The matrix allows global measures of user, producer and overall accuracies to be generated, but lacks consideration of any spatial aspects of accuracy. It is well known that land cover errors are typically spatially auto-correlated and can have a distinct spatial distribution. As yet little work has considered the temporal dimension and investigated the persistence or errors in both geographic and temporal dimensions. Spatio-temporal errors can have a profound impact on both change detection and on environmental monitoring and modelling activities using land cover data. This study investigated methods for describing the spatio-temporal characteristics of classification accuracy. Annual thematic maps were created using a random forest classification of MODIS data over the Jakarta metropolitan areas for the period of 2001–2013. A logistic geographically weighted model was used to estimate annual spatial measures of user, producer and overall accuracies. A principal component analysis was then used to extract summaries of the multi-temporal accuracy. The results showed how the spatial distribution of user and producer accuracy varied over space and time, and overall spatial variance was confirmed by the principal component analysis. The results indicated that areas of homogeneous land cover were mapped with relatively high accuracy and low variability, and areas of mixed land cover with the opposite characteristics. A multi-temporal spatial approach to accuracy is shown to provide more informative measures of accuracy, allowing map producers and users to evaluate time series thematic maps more comprehensively than a standard confusion matrix approach. The need to identify suitable properties for a temporal kernel are discussed.     

Thematic mapping from remotely sensed data with neural networks: MLP, RBF and PNN based approaches

 Neural networks are attractive tools for the derivation of thematic maps from remotely sensed data. Most attention has focused on the multilayer perceptron (MLP) network but other network types are available and have different properties that may sometimes be more appropriate for some applications. Here a MLP, radial basis function (RBF) and probabilistic neural network (PNN) were used to classify remotely sensed data of an agricultural site. The accuracy of these classifications ranged from 86.25–91.25%. The accuracy of the PNN classification could be increased through the incorporation of prior probabilities of class membership but the accuracy of each classification could also be degraded by the presence of an untrained class. Post-classification analyses, however, could be used to identify potentially misclassified cases, including those belonging to an untrained class, to increase accuracy. The effect of the post-classification analysis on the accuracy of the classification derived from each of the three network types investigated differed and it is suggested that network type be selected carefully to meet the requirements of the application in-hand. Received: 23 March 2000 / Accepted: 9 July 2000     

Area-based and location-based validation of classified image objects

Geographic object-based image analysis (GEOBIA) produces results that have both thematic and geometric properties. Classified objects not only belong to particular classes but also have spatial properties such as location and shape. Therefore, any accuracy assessment where quantification of area is required must (but often does not) take into account both thematic and geometric properties of the classified objects. By using location-based and area-based measures to compare classified objects to corresponding reference objects, accuracy information for both thematic and geometric assessment is available. Our methods provide location-based and area-based measures with application to both a single-class feature detection and a multi-class object-based land cover analysis. In each case the classification was compared to a GIS layer of associated reference data using randomly selected sample areas. Error is able to be pin-pointed spatially on per-object, per class and per-sample area bases although there is no indication whether the errors exist in the classification product or the reference data. This work showcases the utility of the methods for assessing the accuracy of GEOBIA derived classifications provided the reference data is accurate and of comparable scale.     

Crop classification accuracy as influenced by training strategy,data transformation and spatial resolution of data

The possibility of improving classification accuracies using different training strategies and data transformations within the framework of a supervised maximum likelihood classification scheme was explored in this study. The effect of spatial resolution of data on the accuracy of classification was also studied Single-pixel training strategy resulted in improved classification accuracy over the block-training method. Data transformations gave no significant improvements in accuracy over untransformed data. There was a reduction in classification accuracy as resolution of data improved from 72 m (LISS I) to 36 m (LISS II) while other sensor characteristics remained same.     

引导型专题数据分级处理研究

通过对已有的数据分级模型及分级模式进行分析,提出引导型专题数据分级处理流程。首先运用分级数确定模型给出分级数,再利用"秩和比法+熵值法"对系统中各个分级模型的分级结果进行评价,最后以向用户推荐分级数和分级模型的形式辅助用户选择合适的分级处理模型。结合实例,对引导型专题数据分级处理过程及关键步骤进行详细说明,并通过原型系统显示实例结果。结果表明,此数据分级处理模式能让普通用户快速准确地进行数据处理。     

Classifying Human Activity Patterns from Smartphone Collected GPS data: A Fuzzy Classification and Aggregation Approach

Smartphones have emerged as a promising type of equipment for monitoring human activities in environmental health studies. However, degraded location accuracy and inconsistency of smartphone‐measured GPS data have limited its effectiveness for classifying human activity patterns. This study proposes a fuzzy classification scheme for differentiating human activity patterns from smartphone‐collected GPS data. Specifically, a fuzzy logic reasoning was adopted to overcome the influence of location uncertainty by estimating the probability of different activity types for single GPS points. Based on that approach, a segment aggregation method was developed to infer activity patterns, while adjusting for uncertainties of point attributes. Validations of the proposed methods were carried out based on a convenient sample of three subjects with different types of smartphones. The results indicate desirable accuracy (e.g. up to 96% in activity identification) using of this method. Two examples are provided in the Appendix to illustrate how the proposed methods could be applied in environmental health studies. Researchers could tailor this scheme to fit a variety of research topics.     

基于随机森林算法的城市不透水面信息提取——以长春市为例

为了快速、准确地掌握不透水面的空间分布及满足动态变化信息现实需求，本文基于多分类器集成学习的思想，引入随机森林算法，以Landsat8影像为数据源，长春市为实验区，选取光谱特征、纹理测度、空间变换后的独立分量等25个特征变量进行分类研究，根据OOB误差进行重要性分析并试验得出最优的分类模型，实现高精度不透水面信息的提取，最后与传统参数分类法进行比较。结果表明：随机森林算法的总体精度可以达到94%，高出最大似然分类法5.9%，支持向量机算法0.77%，Kappa系数为0.914 3，均方根误差为0.104 3，不透水面的提取精度达95.54%，可以精确地得出所需信息，为城市建设与规划提供有效的专题数据。     

光谱角法在ASTER影像土地利用分类中的应用

为了验证光谱角法(SAM)对ASTER影像分类效果,本文对SAM分类的原理进行了阐述和分析,采用了SAM方法以ASTER遥感影像数据为数据源对泸沽湖地区的土地利用进行分类研究,并对分类精度进行了分析。研究结果表明SAM方法用于ASTER数据是一种有效的分类方法,对提高ASTER影像分类精度具有重要的意义。     

遥感图像分类精度的点、群样本检验与评估

遥感专题分类结果在使用前,必须进行客观可靠的精度验证和分析,以保持遥感分类结果的可靠性.本文利用不同分辨率遥感数据获取的同一地区土地利用/覆盖信息,进行了简单随机抽样、系统抽样和分层抽样三种不同抽样组织方式下的点样本和群样本检验分析,评估了不同抽样方式下的点样本和群样本检验效果.研究结果表明：（1）抽样方式对遥感分类精度评价结果的影响是客观存在的,不同抽样方式下的点样本和群样本检验结果都存在一定的随机性,但同一种抽样方式下,点样本检验精度评估结果的波动幅度小于群样本检验,稳定性比群样本检验要好;（2）不同抽样方式下的多次点样本和群样本检验的平均精度检验结果基本上都能够反映分类图像的精度特征,其中,点样本检验中,分层随机抽样点样本检验效果较好;群样本检验中,系统抽样群样本检验和分层随机抽样群样本检验的效果优于简单随机抽样群样本检验.     

Some issues related with sub-pixel classification using HYSI data from IMS-1 satellite

The mixed pixels are treated as noise or uncertainty in class allocation of a pixel and conventional hard classification algorithms may thus produce inaccurate classification outputs. Thus application of sub-pixel or soft classification methods have been adopted for classification of images acquired in complex and uncertain environment. The main objective of this research work has been to study the effect of feature dimensionality using statistical learning classifier — support vector machine (SVM with sigmoid kernel) while using different single and composite operators in fuzzy-based error matrixes generation. In this work mixed pixels have been used at allocation and testing stages and sub-pixel classification outputs have been evaluated using fuzzy-based error matrixes applying single and composite operators for generating matrix. As subpixel accuracy assessment were not available in commercial software, so in-house SMIC (Sub-pixel Multispectral Image Classifier) package has been used. Data used for this research work was from HySI sensor at 506 m spatial resolution from Indian Mini Satellite-1 (IMS-1) satellite launched on April 28, 2008 by Indian Space Research Organisation using Polar Satellite Launch Vehicle (PSLV) C9, acquired on 18^th May 2008 for classification output and IRS-P6, AWIFS data for testing at sub-pixel reference data. The finding of this research illustrate that the uncertainty estimation at accuracy assessment stage can be carried while using single and composite operators and overall maximum accuracy was achieved while using 40 (13 to 52 bands) band data of HySI (IMS-1).     

Dimensionality reduction of hyperspectral data using spectral fractal feature

A new approach for dimensionality reduction of hyperspectral data has been proposed in this article. The method is based on extraction of fractal-based features from the hyperspectral data. The features have been generated using spectral fractal dimension of the spectral response curves (SRCs) after smoothing, interpolating and segmenting the curves. The new features so generated have then been used to classify hyperspectral data. Comparing the post classification accuracies with some other conventional dimensionality reduction methods, it has been found that the proposed method, with less computational complexity than the conventional methods, is able to provide classification accuracy statistically equivalent to those from conventional methods.     

Does environmental data increase the accuracy of land use and land cover classification?

Optical image classification converts spectral data into thematic information from the spectral signature of each object in the image. However, spectral separability is influenced by intrinsic characteristics of the targets, as well as the characteristics of the images used. The classification process will present more reliable results when aspects associated with natural environments (climate, soil, relief, water, etc.) and anthropic environments (roads, constructions, urban area) begin to be considered, as they determine and guide land use and land cover (LULC). The objectives of this study are to evaluate the integration of environmental variables with spectral variables and the performance of the Random Forest algorithm in the classification of Landsat-8 OLI images, of a watershed in the Eastern Amazon, Brazil. The classification process used 96 predictive variables, involving spectral, geological, pedological, climatic and topographic data and Euclidean distances. The selection of variables to construct the predictive models was divided into two approaches: (i) data set containing only spectral variables, and (ii) set of environmental variables added to the spectral data. The variables were selected through nonlinear correlation analysis, with the Randomized Dependence Coefficient and the Recursive Feature Elimination (RFE) method, using the Random Forest classifier algorithm. The spectral variables NDVI, bands 2, 4, 5, 6 and 7 of the dry season and band 4 of the rainy season were selected in both approaches (i and ii). The Euclidean distance from the urban area, Arenosol soil class, annual precipitation, precipitation in February and precipitation of the wettest quarter were the variables selected from the auxiliary data set. This study showed that the addition of environmental data to the spectral data reduces the limitation of the latter, regarding the discrimination of the different classes of LULC, in addition to improving the accuracy of the classification. The addition of soil classes to spectral variables provided a reduction in errors for vegetation classification (Evergreen Forest and Cerrado Sensu Stricto), as it was able to inform about nutrient availability and water storage capacity. The study demonstrates that the addition of environmental variables to the spectral variables can be an alternative to improve monitoring in areas of ecotone in Neotropical regions.     

Radar speckle reduction and derived texture measures for land cover/use classification: a case study

This study examined the appropriateness of radar speckle reduction for deriving texture measures for land cover/use classifications. Radarsat-2 C-band quad-polarised data were obtained for Washington, DC, USA. Polarisation signatures were extracted for multiple image components, classified with a maximum-likelihood decision rule and thematic accuracies determined. Initial classifications using original and despeckled scenes showed despeckled radar to have better overall thematic accuracies. However, when variance texture measures were extracted for several window sizes from the original and despeckled imagery and classified, the accuracy for the radar data was decreased when despeckled prior to texture extraction. The highest classification accuracy obtained for the extracted variance texture measure from the original radar was 72%, which was reduced to 69% when this measure was extracted from a 5 × 5 despeckled image. These results suggest that it may be better to use despeckled radar as original data and extract texture measures from the original imagery.     

Improving Classification Accuracy Assessments with Statistical Bootstrap Resampling Techniques

The use of remotely sensed imagery to generate land cover models is common today. Validation of these models typically involves the use of an independent set of ground-truth data that are used to calculate an error matrix resulting in estimates of omission, commission, and overall error. However, each estimate of error contains a degree of uncertainty itself due to: (1) conceptual bias; (2) location/registration and co-registration errors; and (3) variability in the sample sites used to produce and validate the model. In this study, focus was not placed upon describing land cover mapping techniques, but rather the application of bootstrap resampling to improve the characterization of classification error, demonstrate a method to determine uncertainty from sample site variability, and calculate confidence limits using statistical bootstrap resampling of 500 sample sites acquired within a single Landsat 5 TM image. The sample sites represented one of five land cover categories (water, roads, lava, irrigated agriculture, and rangelands), with each category containing 100 samples. The sample set was then iteratively resampled (n = 200) and 65 sites were randomly selected (without replacement) for use as classification training sites, while the balance (n = 35) were used for validation. Imagery was subsequently classified using a maximum likelihood technique and the model validated using a standard error matrix. This classification-validation process was repeated 200 times. Confidence intervals were then calculated using the resulting omission and commission errors. Results from this experiment indicate that bootstrap resampling is an effective method to characterize classification uncertainty and determine the effect of sample bias.     

利用天宫一号高光谱数据识别植被类型的评估研究

天宫一号高光谱数据尚未得到普遍应用,其数据的质量和应用潜力仍在进一步实践求证和挖掘.See5.0数据挖掘工具是一种能够找出训练样本中模式类隐含特征,并可以自动建立决策规则的分类算法,可避免人为建立分类规则的主观性.本文首先通过光谱曲线分析,选择地物光谱分离性最好的波段组合,然后利用See5.0工具生成规则集,再利用规则集对同一幅天宫一号高光谱数据在不同分类级别上进行分类,并利用相同的验证样本进行精度验证.经过光谱分析发现分类不同森林类型的最佳谱段中心波长分别为:655 nm、673 nm、802 nm、866 nm、984 nm,See5.0分类结果表明在同一树种不同生长期及不同亚种的分类级别上,分类精度在45%以下,表现出了一定局限性,但在树种分类级别上,天宫一号数据表现出了高光谱的优越性,分类精度皆在80%以上,植被类型分类级别,分类精度可达到90%以上.