Improving urban classification through fuzzy supervised classification and spectral mixture analysis

In this study, a fuzzy‐spectral mixture analysis (fuzzy‐SMA) model was developed to achieve land use/land cover fractions in urban areas with a moderate resolution remote sensing image. Differed from traditional fuzzy classification methods, in our fuzzy‐SMA model, two compulsory statistical measurements (i.e. fuzzy mean and fuzzy covariance) were derived from training samples through spectral mixture analysis (SMA), and then subsequently applied in the fuzzy supervised classification. Classification performances were evaluated between the ‘estimated’ landscape class fractions from our method and the ‘actual’ fractions generated from IKONOS data through manual interpretation with heads‐up digitizing option. Among all the sub‐pixel classification methods, fuzzy‐SMA performed the best with the smallest total_MAE (MAE, mean absolute error) (0.18) and the largest Kappa (77.33%). The classification results indicate that a combination of SMA and fuzzy logic theory is capable of identifying urban landscapes at sub‐pixel level.     

Using low‐spectral‐resolution images to acquire simulated hyperspectral images

We propose a method to acquire simulated hyperspectral images using low‐spectral‐resolution images. Hyperspectral images provide more spectral information than low‐spectral‐resolution images, because of the additional spectral bands used for data acquisition in hyperspectral imaging. Unfortunately, original hyperspectral images are more expensive and more difficult to acquire. However, some research questions require an abundance of spectral information for ground monitoring, which original hyperspectral images can easily provide. Hence, we need to propose a method to acquire simulated hyperspectral images, when original hyperspectral images are especially necessary. Since low‐spectral‐resolution images are readily available and cheaper, we develop a method to acquire simulated hyperspectral images using low‐spectral‐resolution images. With simulated hyperspectral images, we can acquire more ‘hidden’ information from low‐spectral‐resolution images. Our method uses the principles of pixel‐mixing to understand the compositional relationship of spectrum data to an image pixel, and to simulate radiation transmission processes. To this end, we use previously obtained data (i.e. spectrum library) and the sorting data of objects that are derived from a low‐spectral‐resolution image. Using the simulation of radiation transmission processes and these different data, we acquire simulated hyperspectral images. In addition, previous analyses of simulated remotely sensed images do not use quantitative statistical measures, but use qualitative methods, describing simulated images by sight. Here, we quantitatively assess our simulation by comparing the correlation coefficients of simulated images and real images. Finally, we use simulated hyperspectral images, real Hyperion images, and their corresponding ALI images to generate several classification images. The classification results demonstrate that simulated hyperspectral data contain additional information not available in the multispectral data. We find that our method can acquire simulated hyperspectral images quickly.     

Mapping of land degradation from space: a comparative study of Landsat ETM+ and ASTER data

The purpose of this study is to compare the role of spectral and spatial resolutions in mapping land degradation from space‐borne imagery using Landsat ETM+ and ASTER data as examples. Land degradation in the form of salinization and waterlogging in Tongyu County, western Jilin Province of northeast China was mapped from an ETM+ image of 22 June 2002 and an ASTER image recorded on 24 June 2001 using supervised classification, together with several other land covers. It was found that the mapping accuracy was achieved at 56.8% and higher for moderately degraded (e.g. salinized) farmland, and over 80% for severely degraded land (e.g. barren) from both ASTER and ETM+ data. The spatial resolution of the ASTER data exerts only a negligible effect on the mapping accuracy. The 30 m ETM+ outperforms the ASTER image of both 15 m and 30 m resolution in consistently generating a higher overall accuracy as well as a higher user's accuracy for barren land. The inferiority of ASTER data is attributed to the highly repetitive spectral content of its six shortwave infrared bands. It is concluded that the spectral resolution of an image is not as important as the information content of individual bands in accurately mapping land covers automatically.     

Parallel processing of remotely sensed hyperspectral imagery: full‐pixel versus mixed‐pixel classification

The rapid development of space and computer technologies allows for the possibility to store huge amounts of remotely sensed image data, collected using airborne and satellite instruments. In particular, NASA is continuously gathering high‐dimensional image data with Earth observing hyperspectral sensors such as the Jet Propulsion Laboratory's airborne visible–infrared imaging spectrometer (AVIRIS), which measures reflected radiation in hundreds of narrow spectral bands at different wavelength channels for the same area on the surface of the Earth. The development of fast techniques for transforming massive amounts of hyperspectral data into scientific understanding is critical for space‐based Earth science and planetary exploration. Despite the growing interest in hyperspectral imaging research, only a few efforts have been devoted to the design of parallel implementations in the literature, and detailed comparisons of standardized parallel hyperspectral algorithms are currently unavailable. This paper compares several existing and new parallel processing techniques for pure and mixed‐pixel classification in hyperspectral imagery. The distinction of pure versus mixed‐pixel analysis is linked to the considered application domain, and results from the very rich spectral information available from hyperspectral instruments. In some cases, such information allows image analysts to overcome the constraints imposed by limited spatial resolution. In most cases, however, the spectral bands collected by hyperspectral instruments have high statistical correlation, and efficient parallel techniques are required to reduce the dimensionality of the data while retaining the spectral information that allows for the separation of the classes. In order to address this issue, this paper also develops a new parallel feature extraction algorithm that integrates the spatial and spectral information. The proposed technique is evaluated (from the viewpoint of both classification accuracy and parallel performance) and compared with other parallel techniques for dimensionality reduction and classification in the context of three representative application case studies: urban characterization, land‐cover classification in agriculture, and mapping of geological features, using AVIRIS data sets with detailed ground‐truth. Parallel performance is assessed using Thunderhead, a massively parallel Beowulf cluster at NASA's Goddard Space Flight Center. The detailed cross‐validation of parallel algorithms conducted in this work may specifically help image analysts in selection of parallel algorithms for specific applications. Copyright © 2008 John Wiley & Sons, Ltd.     

An integrated spatio-temporal classification method for urban fringe change detection analysis

The urban fringe is the transition zone between urban land use and rural land use. It represents the most active part of the urban expansion process. Change detection using multi-temporal imagery is proven to be an efficient way to monitor land-use/land-cover change caused by urban expansion. In this study, we propose a new multi-temporal classification method for change detection in the urban fringe area. The proposed method extracts and integrates spatio-temporal contextual information into multi-temporal image classification. The spatial information is extracted by object-oriented image segmentation. The temporal information is modelled with temporal trajectory analysis with a two-step calibration. A probabilistic schema that employs a global membership function is then used to integrate the spectral, spatial and temporal information. A trajectory accuracy measurement is proposed to assist the comparison on the performances of the integrated spatio-temporal method and classical pixel- and ‘snapshot’-based classification methods. The experiment shows that the proposed method can significantly improve the accuracies of both single scene classification and temporal trajectory analysis.     

Monitoring urban sprawl using simulated PROBA-V data

Urbanization proceeds currently at a rapid pace and the impact on natural ecosystems cannot be neglected. Consequently, it is important to be able to monitor the expansion of urban areas. Yet the process of extracting them from satellite imagery is not trivial. Urban is a non-uniform class with spectral proximity to barren land. In this article, a method for extracting urban areas from medium-resolution Earth observation data is presented. The information source is simulated data of the PROBA-V sensor. Visual and near-infrared bands are classified by the adaptive neuro-fuzzy inference system (ANFIS) neuro-fuzzy classifier into urban and non-urban classes. The method can overcome the main difficulty in similar efforts, i.e. the extensive commission errors of barren to the class urban. The main novelty relies on exploiting annual spectral variability of each land-use class at the pixel level. The basic assumption is that urban and barren areas may have similar spectral values but they have different phenological cycles. The overall accuracy obtained by the classification is 91.57% with a Cohen’s kappa coefficient (khat) of 0.84. Sufficient correlation at the city level is also achieved. Change detection is also possible in terms of hot-spot identification, however marginally suitable for medium-sized cities.     

Land‐cover change and environmental impact analysis in the Greater Mankato area of Minnesota using remote sensing and GIS modelling

Land use and land‐cover (LULC) data provide essential information for environmental management and planning. This research evaluates the land‐cover change dynamics and their effects for the Greater Mankato Area of Minnesota using image classification and Geographic Information Systems (GIS) modelling in high‐resolution aerial photography and QuickBird imagery. Results show that from 1971 to 2003, urban impervious surfaces increased from 18.3% to 32.6%, while cropland and grassland decreased from 54.2% to 39.1%. The dramatic urbanization caused evident environmental impacts in terms of runoff and water quality, whereas the annual air pollution removal rate and carbon storage/sequestration remained consistent since urban forests were steady over the 32‐year span. The results also indicate that highly accurate land‐cover features can be extracted effectively from high‐resolution imagery by incorporating both spectral and spatial information, applying an image‐fusion technique, and utilizing the hierarchical machine‐learning Feature Analyst classifier. This research fills the high‐resolution LULC data gap for the Greater Mankato Area. The findings of the study also provide valuable inputs for local decision‐makers and urban planners.     

Hyperspectral Multiple Features Optimization Using Improved Firefly Algorithm

The utilization of hyperspectral remote sensing image is mainly based on the spectral information,and the spatial information is always be ignored.To solve this problem,a novel hyperspectral multiple features optimization approach based on improved firefly algorithm is presented.Firstly,four spatial features,the local statistical features,gray level co-occurrence matrix features,Gabor filtering features and morphological features of hyperspectral remote sensing image are extracted,and some spectral bands are selected and then combined with these spatial features,and the feature set is constructed.Then,the firefly algorithm is used to optimize the extracted features.In view of the slow convergence speed of firefly algorithm,we use the random inertia weight from particle swarm optimization algorithm to modifiy the location update formula of firefly algorithm,and JM(Jeffreys-Matusita)distance and Fisher Ratio are used as the objective function.Two urban hyperspectral datasets are used for performance evaluation,and the classification results derived from spectral information and spectral-spatial information are compared.The experiments show that random inertia weight can improve the speed of FA-based feature selection algorithm,the performance with multiple features is better than that of spectral information for urban land cover classification,The statistical results of the two sets of experimental data indicate that the selected number of morphological features are the most in the four spatial features.The local statistical features and morphological features are more helpful to the classification of hyperspectral remote sensing images than GLCM and Gabor features.     

A patch‐based image classification by integrating hyperspectral data with GIS

Hyperspectral remote sensing data provide detailed spectral information and are widely used for pixel‐based image classification. However, without considering spatial correlation among neighbouring pixels, a generated thematic map may have a ‘salt‐and‐pepper’ appearance. With the development of the Geographic Information System (GIS), the spatial relationship between a pixel and its neighbours can be recorded readily and used together with remote sensing data. The objective of this study was to integrate hyperspectral data with the GIS for effective thematic mapping. To date, GIS data have been used mainly in field surveys or training field selection for remote sensing data interpretation. Here we propose a patch‐classification based on integration of the GIS with remote sensing data. The classification results obtained by using this method can be easily saved in a vector format as used for GIS files. Computational cost is decreased compared with a pixel‐by‐pixel classification. The issue of how to identify pure or mixed patches is addressed and a three‐level simple and effective checking method is developed. A case study is presented with a hyperspectral data set recorded by the Pushbroom Hyperspectral Imager (PHI) and related GIS data.     

城市建筑材质—地表温度关系的多源遥感研究

利用PROBA CHRIS遥感影像对北京城市建筑材质和自然地表进行基于光谱先验知识的分层分类提取,并与Landsat5 TM热红外数据反演得到的北京城市地表温度叠加,采用统计学方法定量分析了主要建筑材质、自然地类与地表温度的关系,并重点就不同建筑材质对城市热岛的影响及其表面特性所起作用进行了分析。结果表明：北京城区中的砖瓦房表面温度最高,比其他材质高0.3K~4.0K,比自然地类高5.1K~7.8K;金属结构表面温度略低;混凝土、水泥和沥青的平均温度相当,他们是城市热环境异常的主要来源之一;另外,城市中的玻璃幕墙能够有效地降低其表面温度,比其它材质低3.3K~4.0K。反照率、热惯量和热传导性是建筑材质影响城市地表温度的3个重要表面特性,对于不同材质,它们存在较大差异。     

Comparison of pixel‐based and object‐oriented image classification approaches—a case study in a coal fire area,Wuda, Inner Mongolia,China

Pixel‐based and object‐oriented classifications were tested for land‐cover mapping in a coal fire area. In pixel‐based classification a supervised Maximum Likelihood Classification (MLC) algorithm was utilized; in object‐oriented classification, a region‐growing multi‐resolution segmentation and a soft nearest neighbour classifier were used. The classification data was an ASTER image and the typical area extent of most land‐cover classes was greater than the image pixels (15 m). Classification results were compared in order to evaluate the suitability of the two classification techniques. The comparison was undertaken in a statistically rigorous way to provide an objective basis for comment and interpretation. Considering consistency, the same set of ground data was used for both classification results for accuracy assessment. Using the object‐oriented classification, the overall accuracy was higher than the accuracy obtained using the pixel‐based classification by 36.77%, and the user’s and producer’s accuracy of almost all the classes were also improved. In particular, the accuracy of (potential) surface coal fire areas mapping showed a marked increase. The potential surface coal fire areas were defined as areas covered by coal piles and coal wastes (dust), which are prone to be on fire, and in this context, indicated by the two land‐cover types ‘coal’ and ‘coal dust’. Taking into account the same test sites utilized, McNemar’s test was used to evaluate the statistical significance of the difference between the two methods. The differences in accuracy expressed in terms of proportions of correctly allocated pixels were statistically significant at the 0.1% level, which means that the thematic mapping result using object‐oriented image analysis approach gave a much higher accuracy than that obtained using the pixel‐based approach..     

Repeatability of hyperspectral imaging systems—quantification and improvement

This paper addresses a generic problem in remote sensing by aerial hyperspectral imaging systems, that is, very low spatial and spectral repeatability of image cubes. Most analysts are either unaware of this problem or just ignore it. Hyperspectral image cubes acquired in consecutive flights over the same target should ideally be identical. In practice, two consecutive flights over the same target usually yield significant differences between the image cubes. These differences are due to variations in: target characteristics, solar illumination, atmospheric conditions and errors of the imaging system proper. Manufacturers of remote sensing imaging systems use sophisticated equipment to accurately calibrate their instruments, using optimal illumination and constant environment conditions. From a user's perspective, these calibration procedures are only of marginal interest because repeatability is ‘target dependent’. The analyst of hyperspectral imagery is primarily interested in the reliability of the end product, i.e. the repeatability of two image cubes consecutively acquired over the same target, after radiometric calibration, geo‐referencing and atmospheric corrections. Clearly, when the non‐repeatability variance is similar in magnitude to the variance of the spectral or spatial information of interest, it would be impossible to use it for classification or quantification prediction modelling. We present a simple approach for objective assessment of spatial and spectral repeatability by multiple image cube acquisitions, wherein the imaging system views a barium sulphate (BaSO₄) painted panel illuminated by a halogen lamp and by consecutive flights over a reference target. The data analysis is based on several indexes, which were developed for quantifying the spectral and spatial repeatability of hyperspectral image cubes and for detecting outlier voxels. The spectral repeatability information can be used to average less repeatable spectral bands or to exclude them from the analysis. The spatial repeatability information may be used for identifying less repeatable regions of the target. Outlier voxels should be excluded from the analysis because they are grossly erroneous data. Modus operandi for image cube acquisitions is provided, whereby the repeatability may be improved. Spatial and spectral averaging algorithms and software were developed for increasing the repeatability of image cubes in post‐processing.     

Noise reduction of hyperspectral data using singular spectral analysis

In this study, a new noise reduction algorithm based on singular spectral analysis (SSA) was developed to reduce the noise in hyperspectral data. With this SSA‐based approach, the reflectance spectrum of a given pixel in a hyperspectral cube is transformed into its state space. The state space is dynamically constructed and characterized by irregular bases, which allows the proposed approach to reduce noises while keeping the absorption features of surface objects. The performance of the developed method was verified on three datasets: two simulated reflectance spectra with several narrow absorption features and a CHRIS (Compact High Resolution Imaging Spectrometer) data cube over agricultural fields. Our results demonstrated the effectiveness of the SSA‐based approach in improving the signal‐to‐noise ratio of hyperspectral data, while keeping the ‘sharp features’ in the reflectance spectra. The results also show that the proposed SSA method outperforms the commonly used MNF (minimum noise fraction) and wavelet‐based noise reduction methods and it improved vegetation cover classification accuracy by 6%.     

Object-based urban detailed land cover classification with high spatial resolution IKONOS imagery

Improvement in remote sensing techniques in spatial/spectral resolution strengthens their applicability for urban environmental study. Unfortunately, high spatial resolution imagery also increases internal variability in land cover units and can cause a ‘salt-and-pepper’ effect, resulting in decreased accuracy using pixel-based classification results. Region-based classification techniques, using an image object (IO) rather than a pixel as a classification unit, appear to hold promise as a method for overcoming this problem. Using IKONOS high spatial resolution imagery, we examined whether the IO technique could significantly improve classification accuracy compared to the pixel-based method when applied to urban land cover mapping in Tampa Bay, FL, USA. We further compared the performance of an artificial neural network (ANN) and a minimum distance classifier (MDC) in urban detailed land cover classification and evaluated whether the classification accuracy was affected by the number of extracted IO features. Our analysis methods included IKONOS image data calibration, data fusion with the pansharpening (PS) process, Hue–Intensity–Saturation (HIS) transferred indices and textural feature extraction, and feature selection using a stepwise discriminant analysis (SDA). The classification results were evaluated with visually interpreted data from high-resolution (0.3 m) digital aerial photographs. Our results indicate a statistically significant difference in classification accuracy between pixel- and object-based techniques; ANN outperforms MDC as an object-based classifier; and the use of more features (27 vs. 9 features) increases the IO classification accuracy, although the increase is statistically significant for the MDC but not for the ANN.     

Integrating imaging spectroscopy (445–2543 nm) and geographic information systems for post-disaster management: a case of hailstorm damage in Sydney

This paper demonstrates a methodology for the analysis and integration of airborne hyperspectral sensor data (445–2543?nm) with GIS data in order to develop a vulnerability map which has the potential to assist in decision making during post-disaster emergency operations. Hailstorms pose a threat to people as well as property in Sydney, Australia. Emergency planning demands current, large-scale spatio-temporal information on urban areas that may be susceptible to hailstones. Several regions, dominated by less resistant roofing materials, have a higher vulnerability to hailstorm damage than others. Post-disaster operations must focus on allocating dynamic resources to these areas. Remote sensing data, particularly airborne hyperspectral sensor data, consist of spectral bands with narrow bandwidths, and have the potential to quantify and distinguish between urban features such as roofing materials and other man-made features. A spectral library of surface materials from urban areas was created by using a full range spectroradiometer. The image was atmospherically corrected using the empirical line method. A spectral angle mapper (SAM) method, which is an automated method for comparing image spectra to laboratory spectra, was used to develop a classification map that shows the distribution of roofing materials with different resistances to hailstones. Surface truthing yielded high percentage accuracy. Spatial overlay technique was performed in a GIS environment where several types of cartographic data such as special hazard locations, population density, data about less mobile people and the street network were overlaid on the classified geo-referenced hyperspectral image. The integrated database product, which merges high quality spectral information and cartographic GIS data, has vast potential to assist emergency organizations, city planners and decision makers in formulating plans and strategies for resource management.     

结合面向对象方法的多层次干旱与半干旱区域城镇提取方法

在干旱与半干旱区域戈壁及沙漠等高亮地表与城镇连成一片,两者的光谱特征在中等分辨率遥感数据上非常相似;因此,利用基于像素的分类方法很难将城镇准确提取出来。根据两种地物的样本对NDVI、NDBI的分布特征统计分析得出:基于面向对象的分类方法在提取城镇信息方面有较大优势。以典型的干旱区域—黑河流域张掖市及周边地区作为研究区域,将面向对象的方法应用到具有中等分辨率的Landsat-TM数据上,提出了结合面向对象方法的多层次干旱与半干旱区域城镇提取方法。该方法首先使用分层分类的方法得到城镇和荒漠的混合影像,然后使用面向对象的分类方法精确提取城镇信息,其中分割对象过程中引入样本可分离度量化不同尺度的影像分割效果,实现最优尺度分割。结果表明:其目视效果、总体精度(94.51%)和Kappa系数(0.89),均优于支持向量机(SVM)与基于时间序列的分类方法。     

基于光谱特征的城市人工地物分级分类方法研究

１９９７年６月利用ＰＨＩ推帚式成像光谱仪在北京市沙河镇进行了飞行实验,其主要目的是成像光谱技术应用于城市用地和建筑和分类中关键技术研究。对所获取的１５个波段的可见光－近红外数据进行了分析与处理,试图通过对地面覆盖物质的光谱响应特征的分析来进行城市人工目标的识别与分类。由于城市地物的光谱特征异常复杂,难于应用一般的模式分类算法分离出所有类型,本文在对地面覆盖类型光谱响应分析的基础上,采用分层复合分类     

Hyperspectral classification via deep networks and superpixel segmentation

This article presents a new hyperspectral image classification method, which is capable of automatic feature learning while achieving high classification accuracy. The method contains the following two major modules: the spectral classification module and the spatial constraints module. Spectral classification module uses a deep network, called ‘Stacked Denoising Autoencoders’ (SdA), to learn feature representation of the data. Through SdA, the data are projected non-linearly from its original hyperspectral space to some higher-dimensional space, where more compact distribution is obtained. An interesting aspect of this method is that it does not need any prior feature design/extraction process guided by human. The suitable feature for the classification is learnt by the deep network itself. Superpixel is utilized to generate the spatial constraints for the refinement of the spectral classification results. By exploiting the spatial consistency of neighbourhood pixels, the accuracy of classification is further improved by a big margin. Experiments on the public data sets have revealed the superior performance of the proposed method.     

An integrated soft and hard classification approach for evaluating urban expansion from multisource remote sensing data: a case study of the Beijing–Tianjin–Tangshan metropolitan region,China

Integrating soft and hard classification to monitor urban expansion can effectively provide comprehensive urban growth information to urban planners. In this study, both the impervious surface coverage (as a soft classification result) and land cover (as a hard classification result) in the Beijing–Tianjin–Tangshan metropolitan region (BTTMR), China, were extracted from multisource remote sensing data from 1990 to 2015. Then, we evaluated urban expansion based on centre migration, standard deviation ellipse, and spatial autocorrelation metrics. Furthermore, the differences between the soft and hard classification results were analysed at the landscape scale. The results showed that (1) the impervious surface area increased considerably over the past 25 years. Notably, the areas of urban built-up land and industrial production land increased rapidly, while those of ecological land and agricultural production land seriously decreased. (2) The distribution of impervious surfaces was closely related to the regional economic development plan of ‘One Axis, Two Wing, and Multi-Node’ in the BTTMR. (3) The contributions of different land use types to impervious surface growth ranked from high to low as follows: urban built-up land, rural residential land, industrial production land, agricultural production land, and ecological land. (4) The landscape metrics varied considerably based on the hard and soft classification results and were sensitive to different factors.