Classification of multifrequency polarimetric SAR imagery using adynamic learning neural network

A practical method for extracting microwave backscatter for terrain-cover classification is presented. The test data are multifrequency (P, L, C bands) polarimetric SAR data acquired by JPL over an agricultural area called “Flevoland”. The terrain covers include forest, water, bare soil, grass, and eight other types of crops. The radar response of crop types to frequency and polarization states were analyzed for classification based on three configurations: 1) multifrequency and single-polarization images; 2) single-frequency and multipolarization images; and 3) multifrequency and multipolarization images. A recently developed dynamic learning neural network was adopted as the classifier. Results show that using partial information, P-band multipolarization images and multiband hh polarization images have better classification accuracy, while with a full configuration, namely, multiband and multipolarization, gives the best discrimination capability. The overall accuracy using the proposed method can be as high as 95% with a total of thirteen cover types classified. Further reduction of the data volume by means of correlation analysis was conducted to single out the minimum data channels required. It was found that this method efficiently reduces the data volume while retaining highly acceptable classification accuracy     

Genetic Sequence Classification and its Application to Cross-Species Homology Detection

Although large-scale classification studies of genetic sequence data are in progress around the world, very few studies compare different classification approaches, e.g. unsupervised and supervised, in terms of objective criteria such as classification accuracy and computational complexity. In this paper, we study such criteria for both unsupervised and supervised classification of a relatively large sequence data set. The unsupervised approach involves use of different sequence alignment algorithms (e.g., Smith-Waterman, FASTA and BLAST) followed by clustering using the Maximin algorithm. The supervised approach uses a suitable numeric encoding (relative frequencies of tuples of nucleotides followed by principal component analysis) which is fed to a Multi-layer Backpropagation Neural Network. Classification experiments conducted on IBM-SP parallel computers show that FASTA with unsupervised Maximin leads to best trade-off between accuracy and speed among all methods, followed by supervised neural networks as the second best approach. Finally, the different classifiers are applied to the problem of cross-species homology detection.     

认知雷达对抗中的未知雷达状态识别方法

相比于传统的雷达对抗系统,认知雷达对抗引入了闭环行为学习过程,使得干扰方可以通过对雷达信号进行状态辨识,进而进行干扰效果评估,经过自主学习优化干扰策略,从而使得干扰更具有主动性和针对性。雷达状态识别是认知雷达对抗的基础,而在对抗过程中,目标雷达随时可能激活先前“隐藏”的“未知状态”,这就要求干扰方能够快速对未知雷达状态做出响应。本文重点研究认知雷达对抗中的未知雷达状态识别,利用机器学习理论相关算法,提出了基于有监督分类与基于无监督聚类的2种未知状态识别方法,并通过仿真实验分别验证了2种方法的有效性。     

Radar Image Processing for Rock-Type Discrimination

Radar images have unique radiometric and geometric characteristics which present unique problems and opportunities for geological application. This paper reviews preprocessing and analytical techniques found useful or promising for applications of radar images to geologic problems such as rock-type discrimination. The use of coherent monochromatic illumination in radar images results in image speckle noise which interferes with characterization of the imaged surface. Median value filtering of the radar images removes speckle with minimal edge effects and resolution degradation. Variations in radar scene illumination due to uncompensated sensor platform motions or antenna pattern effects can be somewhat corrected for by mean and variance equalization in a direction perpendicular to the resulting image gradient. Registration of radar images to a map base and compensation of terrain induced image distortion can be accomplished by registration to digital elevation models and knowledge of imaging geometry. Analysis of SEASAT images with coregistered LANDSAT images indicates that the radar data can make a significant contribution to rock-type discrimination, especially if textural measures are incorporated. The sensitivity of radar backscatter to local slopes makes radar images an excellent medium from which to extract textural measures. Three techniques for extraction of the textural data inherent in the radar images are presented. Computation of image tone variance over various areas can numerically encode image texture. Hue-saturation-intensity split spectrum processing displays low-frequency variations in color while preserving high-frequency detail.     

Supervised classification of marble textures using support vector machines

A novel approach is described for the supervised classification of marble textures in different classes according to visual appearance, using sum and difference histograms for texture analysis and feature extraction, and support vector machines for classification. Results show very good discrimination between classes.     

Application of neural networks to radar image classification

A number of methods have been developed to classify ground terrain types from fully polarimetric synthetic aperture radar (SAR) images, and these techniques are often grouped into supervised and unsupervised approaches. Supervised methods have yielded higher accuracy than unsupervised techniques, but suffer from the need for human interaction to determine classes and training regions. In contrast, unsupervised methods determine classes automatically, but generally show limited ability to accurately divide terrain into natural classes. In this paper, a new terrain classification technique is introduced to determine terrain classes in polarimetric SAR images, utilizing unsupervised neural networks to provide automatic classification, and employing an iterative algorithm to improve the performance. Several types of unsupervised neural networks are first applied to the classification of SAR images, and the results are compared to those of more conventional unsupervised methods. Results show that one neural network method-Learning Vector Quantization (LVQ)-outperforms the conventional unsupervised classifiers, but is still inferior to supervised methods. To overcome this poor accuracy, an iterative algorithm is proposed where the SAR image is reclassified using a maximum likelihood (ML) classifier. It is shown that this algorithm converges, and significantly improves classification accuracy     

A Markov random field model for classification of multisourcesatellite imagery

A general model for multisource classification of remotely sensed data based on Markov random fields (MRF) is proposed. A specific model for fusion of optical images, synthetic aperture radar (SAR) images, and GIS (geographic information systems) ground cover data is presented in detail and tested. The MRF model exploits spatial class dependencies (spatial context) between neighboring pixels in an image, and temporal class dependencies between different images of the same scene. By including the temporal aspect of the data, the proposed model is suitable for detection of class changes between the acquisition dates of different images. The performance of the proposed model is investigated by fusing Landsat TM images, multitemporal ERS-1 SAR images, and GIS ground-cover maps for land-use classification, and on agricultural crop classification based on Landsat TM images, multipolarization SAR images, and GIS crop field border maps. The performance of the MRF model is compared to a simpler reference fusion model. On an average, the MRF model results in slightly higher (2%) classification accuracy when the same data is used as input to the two models. When GIS field border data is included in the MRF model, the classification accuracy of the MRF model improves by 8%. For change detection in agricultural areas, 75% of the actual class changes are detected by the MRF model, compared to 62% for the reference model. Based on the well-founded theoretical basis of Markov random field models for classification tasks and the encouraging experimental results in our small-scale study, the authors conclude that the proposed MRF model is useful for classification of multisource satellite imagery     

Classification of Convective and Stratiform Cells in Meteorological Radar Images Using SVM Based on a Textural Analysis

This contribution deals with the discrimination between stratiform and convective cells in meteorological radar images. This study is based on a textural analysis of the latter and their classification using a support vector machine (SVM). First, we apply different textural parameters such as energy, entropy, inertia, and local homogeneity. Through this experience, we identify the different textural features of both the stratiform and convective cells. Then, we use an SVM to find the best discriminating parameter between the two types of clouds. The main goal of this work is to better apply the Palmer and Marshall Z-R relations specific to each type of precipitation.     

Application of neural networks for sea ice classification inpolarimetric SAR images

Several automatic methods have been developed to classify sea ice types from fully polarimetric synthetic aperture radar (SAR) images, and these techniques are generally grouped into supervised and unsupervised approaches. In previous work, supervised methods have been shown to yield higher accuracy than unsupervised techniques, but suffer from the need for human interaction to determine classes and training regions. In contrast, unsupervised methods determine classes automatically, but generally show limited ability to accurately divide terrain into natural classes. In this paper, a new classification technique is applied to determine sea ice types in polarimetric and multifrequency SAR images, utilizing an unsupervised neural network to provide automatic classification, and employing an iterative algorithm to improve the performance. The learning vector quantization (LVQ) is first applied to the unsupervised classification of SAR images, and the results are compared with those of a conventional technique, the migrating means method. Results show that LVQ outperforms the migrating means method, but performance is still poor. An iterative algorithm is then applied where the SAR image is reclassified using the maximum likelihood (ML) classifier. It is shown that this algorithm converges, and significantly improves classification accuracy. The new algorithm successfully identifies first-year and multiyear sea ice regions in the images at three frequencies. The results show that L- and P-band images have similar characteristics, while the C-band image is substantially different. Classification based on single features is also carried out using LVQ and the iterative ML method. It is found that the fully polarimetric classification provides a higher accuracy than those based on a single feature. The significance of multilook classification is demonstrated by comparing the results obtained using four-look and single-look classifications     

基于聚类识别的极化SAR图像分类

This paper presents a novel unsupervised image classification method for Polarimetric Synthetic Aperture Radar (PolSAR) data. The proposed method is based on a discriminative clustering framework that explicitly relies on a discriminative supervised classification technique to perform unsupervised clustering. To implement this idea, an energy function is designed for unsupervised PolSAR image classification by combining a supervised Softmax Regression (SR) model with a Markov Random Field (MRF) smoothness constraint. In this model, both the pixelwise class labels and classifiers are taken as unknown variables to be optimized. Starting from the initialized class labels generated by Cloude-Pottier decomposition and K-Wishart distribution hypothesis, the classifiers and class labels are iteratively optimized by alternately minimizing the energy function with respect to them. Finally, the optimized class labels are taken as the classification result, and the classifiers for different classes are also derived as a side effect. This approach is applied to real PolSAR benchmark data. Extensive experiments justify that the proposed approach can effectively classify the PolSAR image in an unsupervised way and produce higher accuracies than the compared state-of-the-art methods.     

Method for unsupervised classification of multiunit neural signal recording under low signal-to-noise ratio

Neural spike sorting is an indispensable step in the analysis of multiunit extracellular neural signal recording. The applicability of spike sorting systems has been limited, mainly to the recording of sufficiently high signal-to-noise ratios, or to the cases where supervised classification can be utilized. We present a novel unsupervised method that shows satisfactory performance even under high background noise. The system consists of an efficient spike detector, a feature extractor that utilizes projection pursuit based on negentropy maximization (Huber, 1985 and Hyvarinen et al, 1999), and an unsupervised classifier based on probability density modeling using mixture of Gaussians (Jain et al., 2000). Our classifier is based on the mixture model with a roughly approximated number of Gaussians and subsequent mode-seeking. It does not require accurate estimation of the number of units present in the recording and, thus, is better suited for use in fully automated systems. The feature extraction stage leads to better performance than those utilizing principal component analysis and two nonlinear mappings for the recordings from the somatosensory cortex of rat and the abdominal ganglion of Aplysia. The classification method yielded correct classification ratio as high as 95%, for data where it was only 66% when a kappa-means-type algorithm was used for the classification stage.     

Unsupervised Quick Reduct Algorithm Using Rough Set Theory

Feature selection (FS) is a process to select features which are more informative.It is one of the important steps in knowledge discovery.The problem is that not all features are important.Some of the features may be redundant,and others may be irrelevant and noisy.The conventional supervised FS methods evaluate various feature subsets using an evaluation function or metric to select only those features which are related to the decision classes of the data under consideration.However,for many data mining applications,decision class labels are often unknown or incomplete,thus indicating the significance of unsupervised feature selection.However,in unsupervised learning,decision class labels are not provided.In this paper,we propose a new unsupervised quick reduct (QR) algorithm using rough set theory.The quality of the reduced data is measured by the classification performance and it is evaluated using WEKA classifier tool.The method is compared with existing supervised methods and the result demonstrates the efficiency of the proposed algorithm.     

Integration of spatial and spectral information by means of unsupervised extraction and classification for homogenous objects applied to multispectral and hyperspectral data

This paper presents a method of unsupervised enhancement of pixels homogeneity in a local neighborhood. This mechanism will enable an unsupervised contextual classification of multispectral data that integrates the spectral and spatial information producing results that are more meaningful to the human analyst. This unsupervised classifier is an unsupervised development of the well-known supervised extraction and classification for homogenous objects (ECHO) classifier. One of its main characteristics is that it simplifies the retrieval process of spatial structures. This development is specially relevant for the new generation of airborne and spaceborne sensors with high spatial resolution.     

Classification of Simulated And Actual NOAA-6 AVHRR Data for Hydrologic Land-Surface Feature Definition

Current ground hydrology models (GHM's) require global distribution of bare soil and vegetation, the physical and thermal properties of soil, and the physiological and physical properties of plants to parameterize evaporation and the sensible heat flux from the land surfaces. Thus the ability to infer vegetative cover, and to some extent vegetation type, is hydrologically important because of the relationship between vegetation and evapotranspiration through the process of root-zone soil-moisture extraction. LANDSAT digital data degraded to approximately 1 km and NOAA-6 digital data have been used to study the capability and problems associated with the use of low-resolution data to provide land-surface information such as forest, grassland, agriculture, bare, urban, and water. Three LANDSAT scenes and a subscene from a NOAA-6 pass were classified using supervised and unsupervised techniques. The LANDSAT data were used initially to study classification techniques and ascertain problems associated with large-scale classification prior to the receipt of NOAA-6 data. Comparisons between the LANDSAT supervised classification (?ground truth?) and the unsupervised classification resulted in percentage differences between the cover types of generally less than 10 percent. The Advanced Very Hign Resolution Radiometer (AVHRR) results were similar to the LANDSAT. In both cases there was no statistical difference between the supervised and unsupervised results. The major problem encountered was consistent labeling of the various landcover categories derived by the classification methods.     

Integrated spectral and spatial information mining in remote sensing imagery

Most existing remote sensing image retrieval systems allow only simple queries based on sensor, location, and date of image capture. This approach does not permit the efficient retrieval of useful hidden information from large image databases. This paper presents an integrated approach to retrieving spectral and spatial patterns from remotely sensed imagery using state-of-the-art data mining and advanced database technologies. Land cover information corresponding to spectral characteristics is identified by supervised classification based on support vector machines with automatic model selection, while textural features characterizing spatial information are extracted using Gabor wavelet coefficients. Within identified land cover categories, textural features are clustered to acquire search-efficient space in an object-oriented database with associated images in an image database. Interesting patterns are then retrieved using a query-by-example approach. The evaluation of the study results using coverage and novelty measures validates the effectiveness of the proposed remote sensing image information mining framework, which is potentially useful for applications such as agricultural and environmental monitoring.     

Textural Analysis And Real-Time Classification of Sea-Ice Types Using Digital SAR Data

Digital measures of synthetic-aperture radar (SAR) image texture, as well as the local approximation to the mean value of individual ice types, were used to perform discrimination and mapping of ice types. The SAR data described in this paper were gathered in March, 1979, over the Beaufort Sea as part of the Canadian SURSAT project. Digital SAR data from a 3 × 3 km area were obtained using optical processing of the signal film and digital recording of the output image. Prior to performing the textural analysis, a digital filter algorithm was developed that minimizes the effect of radar-system-generated coherent speckle and produces an image approximating local tone while preserving edge definition. This image was used in the analysis to separate image tone from image texture. The textural analysis, which included calculating the entropy and inertia of the image, indicated that first- and multiyear, smooth- and rough-ice types could be distinguished based on the textural values obtained from the data with an overall accuracy of 65 percent. This study has also considered the use of cellular operations based upon neighborhood transformations to calculate the textural values. This computation method can potentially reduce the time to compute textural features on a general-purpose computer to near real-time rates.     

地基雷达地物检测算法研究

为了提高地基雷达系统的监测精度,提出了一种以散射模型为基础的Freeman-Durden[]分解算法和基于非高斯分布的K-Wishart算法相结合的混合型高斯分布迭代无监督分类算法,对比Freeman-Durden分解和复Wishart分布组合算法,该算法具有更好的分类性能。实验结果表明,该算法不仅适合于Wishart、K-wishart分布对均匀区域数据的描述,而且对一般不均匀区域数据的描述也很强。     

Texture classification with kernel principal component analysis

Kernel principal component analysis (PCA) is presented as a mechanism for extracting textural information. Using the polynomial kernel, higher order correlations of input pixels can be easily used as features for classification. As a result, supervised texture classification can be performed using a neural network     

Detection of daily activities and sports with wearable sensors in controlled and uncontrolled conditions

Physical activity has a positive impact on people's well-being, and it may also decrease the occurrence of chronic diseases. Activity recognition with wearable sensors can provide feedback to the user about his/her lifestyle regarding physical activity and sports, and thus, promote a more active lifestyle. So far, activity recognition has mostly been studied in supervised laboratory settings. The aim of this study was to examine how well the daily activities and sports performed by the subjects in unsupervised settings can be recognized compared to supervised settings. The activities were recognized by using a hybrid classifier combining a tree structure containing a priori knowledge and artificial neural networks, and also by using three reference classifiers. Activity data were collected for 68 h from 12 subjects, out of which the activity was supervised for 21 h and unsupervised for 47 h. Activities were recognized based on signal features from 3-D accelerometers on hip and wrist and GPS information. The activities included lying down, sitting and standing, walking, running, cycling with an exercise bike, rowing with a rowing machine, playing football, Nordic walking, and cycling with a regular bike. The total accuracy of the activity recognition using both supervised and unsupervised data was 89% that was only 1% unit lower than the accuracy of activity recognition using only supervised data. However, the accuracy decreased by 17% unit when only supervised data were used for training and only unsupervised data for validation, which emphasizes the need for out-of-laboratory data in the development of activity-recognition systems. The results support a vision of recognizing a wider spectrum, and more complex activities in real life settings.