Genetic Algorithm Based Feature Subset Selection for Land Cover/ Land Use Mapping Using Wavelet Packet Transform

A genetic algorithm based approach is used in this paper for the selection of a subset from the combination of Wavelet Packet Statistical and Wavelet Packet Co-occurrence textural feature sets to classify the LISS IV satellite images using neural networks. Generally, adding a new feature increases the complexity of training and classification. Hence there is a need to differentiate between those features that contribute ample information and others. Many current feature reduction techniques such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) involve linear transformations of the original pattern vectors to new vectors of lower dimensions. Hence a multi-objective Genetic Algorithm has been employed to reduce the complexity and increase the accuracy of classification. Four indices - user’s accuracy, producer’s accuracy, overall accuracy and kappa co-efficient are used to assess the accuracy of the classified data. Experimental results show that the proposed Genetic Algorithm approach with lesser number of optimal features produces comparable results with that of our earlier approach using more features.     

Land Cover/Land use Mapping Using Different Wavelet Packet Transforms for LISS IV Madurai Imagery

Texture in high resolution satellite images requires substantial improvement in the conventional segmentation algorithms. The use of wavelet packet transforms for texture analysis and image classification of high spatial resolution LISS IV imagery provide more details about the urban areas. This paper analyses the performance of a combination of Wavelet Packet Statistical Features (WPSFs) and Wavelet Packet Co-occurrence Features (WPCFs) for the classification of LISS IV images. The classification accuracy per pixel is improved in this paper by varying the window size. Four indices—user’s accuracy, producer’s accuracy, overall accuracy and kappa co-efficient are used to assess the accuracy of the classified data. Experimental results show that a multi-band and multi-level wavelet packet approach can be used to drastically increase the classification accuracy.     

ALOS PALSAR and Hyperion Data Fusion for Land Use Land Cover Feature Extraction

Detailed and enhanced land use land cover (LULC) feature extraction is possible by merging the information extracted from two different sensors of different capability. In this study different pixel level image fusion algorithms (PCA, Brovey, Multiplicative, Wavelet and combination of PCA & IHS) are used for integrating the derived information like texture, roughness, polarization from microwave data and high spectral information from hyperspectral data. Span image which is total intensity image generated from Advanced Land observing Satellite-Phase array L-band SAR (ALOS-PALSAR) quad polarization data and EO-1 Hyperion data (242 spectral bands) were used for fusion. Overall PCA fused images had shown better result than other fusion techniques used in this study. However, Brovey fusion method was found good for differentiating urban features. Classification using support vector machines was conducted for classifying Hyperion, ALOS PALSAR and fused images. It was observed that overall classification accuracy and kappa coefficient with PCA fused images was relatively better than other fusion techniques as it was able to discriminate various LULC features more clearly.     

面向城市地物分类的 SAR 图像纹理特征提取与分析

合成孔径雷达（ SAR）图像含有丰富的纹理信息,特别是进行城市地物分类时,纹理特征对于图像的解译具有重要的意义。本文对基于灰度共生矩阵和Gabor变换两种纹理特征提取方法进行了研究,将灰度和不同纹理特征组合应用于SAR图像城市地物分类,并以ALOS PALSAR影像为数据源进行了实验。通过对不同分类结果进行定性和定量分析,结果表明,引入纹理特征后的SAR图像分类结果要优于无纹理信息参与的分类结果,基于不同纹理特征组合的SAR图像分类结果要优于基于单一纹理特征的分类结果。     

Gradient-Magnitude-Based Support Regions in Structural Land Use Classification

Land use classification is one of the major problems in remote sensing. Previous studies focused on multispectral information, texture-based features, and features based on edge detection to classify land usage from satellite images. In a previous study, structural features are introduced to classify land development using high-resolution satellite images. These structural features were based on line support regions (LSRs). LSRs are introduced to detect and represent straight lines in images using a pixel-grouping process. The structural features are calculated on these grouped pixels. It is shown that gradient-magnitude-based pixel grouping may also be used in structural feature calculations. Therefore, the aim of this letter is twofold. First, the previous structural feature calculation method is shown to be more general than the LSR. Second, LSR-based features are shown to require fairly high computation compared to gradient-magnitude-based features with similar classification performance     

A Novel Remote Sensing Image Classification Scheme Based on Data Fusion, Multiple Features and Ensemble Learning

In this paper, we propose a novel scheme to improve the accuracy of remote sensing image classification by integrating data fusion, multiple feature combination and ensemble learning. Intensity-Hue-Saturation (IHS), Gram-Schmidt (GS), Brovey and wavelet fusion methods are first performed to obtain the optimal fusion images of high resolution and multispectral images. Support Vector Machine (SVM) classifier is then adopted to classify the fused image with different feature sets, and ensemble learning algorithm based on dynamic classifier selection (DCS) is finally used to integrate multiple classification maps. The proposed classification scheme is implemented with three remote sensing data sets, obtaining the highest overall accuracy and kappa coefficient in all cases (92.63% and 0.8917 for BJ-1 data set, 81.89% and 0.7513 for Landsat TM and SPOT4 data set, 92.21% and 0.8838 for ALOS data set respectively). The experimental results show that the integration of data fusion, feature combination and ensemble learning improves the classification performance obviously and has great potential in practical uses.     

顾及分类器参数的全极化SAR图像特征选择与分类

全极化SAR获取的信息量远多于传统SAR,但信息量的增加并不能确保分类精度的提高,如何有效进行特征选择至关重要。针对自适应特征选择问题,提出一种顾及分类器参数的特征选择和分类方法。该方法以支持向量数为评估依据,结合遗传算法进行特征选择,并同时对分类器参数进行寻优;最后利用优选的特征集和模型参数进行分类。为验证算法的有效性,利用两组全极化数据进行了监督分类实验。实验结果表明,提出方法降低了SVM分类器对自身参数的敏感性,而且能在较少特征个数下具备良好的泛化性能,分类精度优于未经过特征选择和参数优化的方法。     

面向对象方法的时间序列MODIS数据湿地信息提取——以洞庭湖流域为例

以洞庭湖流域为研究区,对大范围湿地信息遥感提取方法进行了研究。先基于时间序列MODIS EVI及物候特征参数,通过J-M(Jeffries-Matusita distance)距离分析,构建了MODIS(250 m)最佳时序组合分类数据;其次,通过Johnson指数确定了最佳分割尺度,采用面向对象的遥感分类方法(Random tree分类器)提取了洞庭湖流域的湿地信息,并验证该方法的适用性。研究结果表明,基于时序数据与面向对象的Random tree分类的总体精度和Kappa系数分别为78.84%和0.71,较之基于像元的相同算法的总体分类精度和Kappa系数分别提高了5.79%和0.04。同时,基于面向对象方法的湿地整体的用户精度与生产者精度较基于像元方法分别提高了4.56%和6.21%,可有效提高大区域湿地信息提取的精度。     

A Hybrid Approach for Land Use/Land Cover Classification

Atlanta has continuously changed its physical landscape as well as its socioeconomic appearance over the past decades. A hybrid image processing approach, which integrated unsupervised, supervised, and spectral mixture analysis (SMA) classification methods, was used to identify urban land use/land cover changes over a decade (from 1990 to 2000) in the Atlanta metropolitan area. During this process, SMA was proven to be an effective analytical approach for characterizing mixed feature areas, such as a metropolitan area. According to accuracy assessment, the classification results were acceptable.     

高光谱遥感影像SVM分类中训练样本选择的研究

支持向量机(SVM)分类的关键是发现分类最优超平面及类别间隔,而混合像元比纯净像元更接近类别边界,更容易找出最优超平面。本文针对SVM分类器的特点,在高光谱数据分类中采用混合像元作为训练样本对SVM进行训练,试验表明采用类别边界上的混合像元作为训练样本是可行的,能够获得与纯净训练样本接近的分类精度,进一步验证了SVM分类对训练样本空间分布依赖度较低的特点。     

基于蚁群优化的特征选择新方法

利用蚁群优化算法解决特征选择问题，以获得能代表问题空间的较优特征子集，并能降低分类系统的搜索空间。以航空纹理影像的特征选择和分类问题为例，利用主分量变换和蚁群优化算法分别对原始纹理影像特征集合进行特征提取、选择和分类。结果表明，本文方法不仅能够降低图像特征空间维数，减少图像分类的工作量，而且还可以提高分类识别的正确率。     

面向地块的MDPA算法实现Landsat7的农用地分类

介绍了使用地块划分Landsat7 ETM+遥感影像作为分类单元。为每一种目标类型选取标准地块,采用MDPA算法计算待分类地块与标准地块的亮度、绿度和湿度3种特征的直方图距离和相似指数,以此作为分类依据,结果准确率为95%。对于地块边界稳定的区域,能够显著地降低工作强度,提高分类精度。     

单变量特征选择的苏北地区主要农作物遥感识别

遥感识别多源特征综合和特征优选是提高遥感影像分类精度的关键技术。农作物遥感识别中,识别特征的相对单一和数量过多均会导致作物识别精度不理想。随机森林(random forests)采用分类与回归树(CART)算法来生成分类树,结合了bagging和随机选择特征变量的优点,是一种有效的分类方法。单变量特征选择(univariate feature selection)能够对每一个待分类的特征进行测试,衡量该特征和响应变量之间的关系,根据得分舍弃不好的特征,优选得到的特征用于分类。本文基于随机森林和单变量特征选择,利用多时相光谱信息、植被指数信息、纹理信息及波段差值信息,设计多组分类实验方案,对江苏省泗洪县的高分一号(GF-1)和环境一号(HJ-1A)影像进行分类研究,旨在选择最佳的分类方案对实验区主要农作物进行识别和提取。实验结果表明:(1)多源信息综合的农作物分类精度明显高于单一的原始光谱特征分类,说明不同类型特征的引入能改善分类效果;(2)基于单变量特征选择算法的优选特征分类效果最佳,总体精度97.07%,Kappa系数0.96,表明了特征优选在降低维度的同时,也保证了较高的分类精度。随机森林和单变量特征选择结合的方法可以提高遥感影像的分类精度,为农作物的识别和提取研究提供了有效的方法。     

Cotton Crop Discrimination Using Fuzzy Classification Approach

Crop growth information represented through temporal remote sensing data is of great importance for specific agriculture crop discrimination. In this paper, the effect of various indices was empirically investigated using temporal images for cotton crop discrimination. Five spectral indices SR (Simple Ratio), NDVI (Normalized Difference Vegetation index), TNDVI (Transformed Normalized Difference Vegetation Index), SAVI (Soil-Adjusted Vegetation Index) and TVI (Triangular Vegetation Index) were investigated to identify cotton crop using temporal multi-spectral images. Data used for this study was AWIFS (coarser resolution) for soft classification and LISS-III (medium coarser) data for soft testing from Resourcesat-1 (IRS-P6) satellite. The mixed pixel (i.e. multiple classes within a single pixel) problem had been handled using soft computing techniques. Possibilistic fuzzy classification approach is used to handle mixed pixels for extracting single class of interest. The classification results with respect to various indices were compared in terms of image to image fuzzy overall classification accuracy. It was observed that temporal SAVI indices database with data set-2 outperformed other temporal indices database for cotton crop discrimination. Temporal SAVI indices database gave highest fuzzy overall accuracy of 93.12% with data set-2 in comparison to others.     

Exploitation of TerraSAR-X Data for Land use/Land Cover Analysis Using Object-Oriented Classification Approach in the African Sahel Area, Sudan

Recently, object-oriented classification techniques based on image segmentation approaches are being studied using high-resolution satellite images to extract various thematic information. In this study different types of land use/land cover (LULC) types were analysed by employing object-oriented classification approach to dual TerraSAR-X images (HH and HV polarisation) at African Sahel. For that purpose, multi-resolution segmentation (MRS) of the Definiens software was used for creating the image objects. Using the feature space optimisation (FSO) tool the attributes of the TerraSAR-X image were optimised in order to obtain the best separability among classes for the LULC mapping. The backscattering coefficients (BSC) for some classes were observed to be different for HH and HV polarisations. The best separation distance of the tested spectral, shape and textural features showed different variations among the discriminated LULC classes. An overall accuracy of 84 % with a kappa value 0.82 was resulted from the classification scheme, while accuracy differences among the classes were kept minimal. Finally, the results highlighted the importance of a combine use of TerraSAR-X data and object-oriented classification approaches as a useful source of information and technique for LULC analysis in the African Sahel drylands.     

面向对象的无人机遥感影像岩溶湿地植被遥感识别

以广西桂林会仙喀斯特国家湿地公园为研究区,以无人机航摄影像为数据源,综合利用面向对象的影像分析技术、随机森林算法、阈值分类方法和Boruta全相关特征变量选择算法进行岩溶湿地植被的遥感识别。结果表明：针对不同特征变量对岩溶湿地遥感识别的贡献率而言,光谱特征（DOM > DSM） > 纹理特征（DOM > DSM） > 几何特征 > 上下文变量;两个航摄影像数据集的总体分类精度都在85%以上,Kappa系数也高于0.85。本文研究结果对基于高空间分辨率无人机可见光影像的岩溶湿地植被遥感识别在特征变量选择、分割参数选择及方法选择方面具有一定的借鉴意义。     

Combining the FCM Classifier with Various Kernels to Handle Non-linearity of Class Boundaries

This article presents the use of kernel functions in fuzzy classifiers for an efficient land use/land cover mapping. It focuses on handling mixed pixels obtained from a remote sensing image by considering non-linearity between class boundaries. It uses kernel functions combined with the conventional fuzzy c-means (FCM) classifier. Kernel-based fuzzy c-mean classifiers were applied to classify AWiFS and LISS-III images from Resourcesat-1 and Resourcesat-2 satellites. Optimal kernels were obtained from eight single kernel functions. Fractional images generated from high resolution LISS-IV image were used as reference data. Classification accuracy of the FCM classifier increased with 12.93%. Improvement in overall accuracy shows that non-linearity in the dataset was handled adequately. The inverse multiquadratic kernel and the Gaussian kernel with the Euclidean norm were identified as optimal kernels. The study showed that overall classification accuracy of the FCM classifier improved if kernel functions were included.     

基于多特征的遥感影像分类方法

提出了一种基于多特征的遥感分类方法。首先 ,制定类方案并分解各个类 ,据此得到相应的子类 ;然后 ,通过选用适当的特征 ,使得每一个类都能以一个独特的特征组合来表达。与此同时 ,通过影像分割得到影像对象 ,并测量这些对象的各个特征 ,如光谱特征、几何特征及拓扑特征等。凭借这些特征 ,影像对象可以较为容易地被识别和分类。与传统的分类方法的比较表明 ,文中所提出的分类方法具有明显的优越性和良好的前景。     

Improving Spectral–spatial Classification of Hyperspectral Imagery Using Spectral Dimensionality Reduction Based on Weighted Genetic Algorithm

This paper aims to improve the accuracy and the efficiency of high resolution land cover mapping in urban area. To this end, an improved approach for classification of hyperspectral imagery is proposed and evaluated. This approach benefits from both inherent spectral and spatial information of an image. The weighted genetic (WG) algorithm is first used to obtain the subspace of hyperspectral data. The obtained features are then fed into the enhanced marker-based minimum spanning forest (EMSF) classification algorithm. In this algorithm, the markers are extracted from the classification maps obtained by both support vector machine and watershed segmentation algorithm classifiers. For this purpose, the class’s pixels with the largest population in the classification map are kept for each region of the segmentation map. Then, the most reliable classified pixels are chosen from among the exiting pixels as markers. To evaluate the efficiency of the proposed approach, three hyperspectral data sets acquired by ROSIS-03, Hymap and Hyper-Cam LWIR are used. Experimental results showed that the proposed WG–EMSF approach achieves approximately 9, 8 and 6% better overall accuracy than the original MSF-based algorithm for these data sets respectively.     

Comparison of Pixel and Object Oriented Based Classification of Hyperspectral Pansharpened Images

In this paper pixel-based and object-oriented classifications were investigated for land-cover mapping in an urban area. Since the image fusion methods are playing a useful role in supplying classification different fusion approaches such as Gram-Schmidt Transform (GS), Principal Component Transform (PC), Haar wavelet, and À Trous Wavelet Transform (ATWT) algorithms have been used and the fused image with the best quality has been assessed on its respected classification. A Hyperion image and IRS-PAN image covering a region near Tehran, Iran have been used to demonstrate the enhancement and accuracy assessment of fused image over the initial images. The evaluation results of fused images showed that the Haar wavelet approach has good quality in preserving spectral information as well as spatial information. Classification results were compared to evaluate the effectiveness of the two classification approaches. Result of the pan-sharpened image classifications displayed that the object-oriented procedure presented more accurate outcomes (90.47 %) than those obtained by pixel-based classification method (77.33 %).