Multi-scale Texture Features Research Articles

Towards an automated monitoring of human settlements in South Africa using high resolution SPOT satellite imagery

Abstract. Urban areas in sub-Saharan Africa are growing at an unprecedented pace. Much of this growth is taking place in informal settlements. In South Africa more than 10% of the population live in urban informal settlements. South Africa has established a National Informal Settlement Development Programme (NUSP) to respond to these challenges. This programme is designed to support the National Department of Human Settlement (NDHS) in its implementation of the Upgrading Informal Settlements Programme (UISP) with the objective of eventually upgrading all informal settlements in the country. Currently, the NDHS does not have access to an updated national dataset captured at the same scale using source data that can be used to understand the status of informal settlements in the country. This pilot study is developing a fully automated workflow for the wall-to-wall processing of SPOT-5 satellite imagery of South Africa. The workflow includes an automatic image information extraction based on multiscale textural and morphological image features extraction. The advanced image feature compression and optimization together with innovative learning and classification techniques allow a processing of the SPOT-5 images using the Landsat-based National Land Cover (NLC) of South Africa from the year 2000 as low-resolution thematic reference layers as. The workflow was tested on 42 SPOT scenes based on a stratified sampling. The derived building information was validated against a visually interpreted building point data set and produced an accuracy of 97 per cent. Given this positive result, is planned to process the most recent wall-to-wall coverage as well as the archived imagery available since 2007 in the near future.

Color and multiscale texture features from vectorial mathematical morphology

In this paper, we explore an original way to compute texture features for color images in a vector process. Using a dedicated approach for color ordering, we produce a complete framework for color mathematical morphology adapted to human visual system characteristics. Then, morphological multiscale texture features are defined. To understand the texture feature behavior, we present the feature response to basic images variations. Finally, we compare the texture feature performance in front of a classical classification task using Outex database.

Texture Feature Extraction of Hyper-spectral Image with Three-dimensional Gray-level Co-occurrence

To extract useful information of hyper-spectral images effectively, a kind of texture feature extraction method using three-dimensional gray-level co-occurrence matrix (3d glcm) is proposed in this paper. the method extracts the texture features of hyper-spectral image as a pseudo data cube that combines the two-dimensional space data with one-dimensional spectrum data, instead of each band computed alone. moreover, the parameters related to building the 3d glcm are all optimized. to obtain the features both in spectral space and spatial space, the moving directions of texture window are extended to the spectral space, namely that four directions in two-dimensional (2d) image space are expanded to thirteen ones in three-dimensional (3d) space. then, the jeffreys-matusita (jm) distance based on the class separable criterion is employed to select the most suitable window size for each object. finally, the multi-scale texture features are used for classification. the experiments also show that, compared with the traditional methods, the feature extraction method is more effective in describing objects and has better classification accuracy. ©, 2015, binary information press

An integrated multiscale and multivariate image analysis framework for process monitoring of colour random textures: MSMIA

We present an integrated approach for conducting on-line and off-line image-based monitoring of processes whose products (raw materials, intermediate or final) consist of colour random textures. The methodology combines the principles underlying wavelet texture analysis and multivariate image analysis into a single framework, able to detect both abnormal changes in texture and colour. By taking into account a scale-dependent description of colour, it can detect subtle changes on how colour interacts with texture across the several length-scales considered. The proposed methodology was studied and characterized following best practice procedures for developing statistical process control methods, where controlled simulated test scenarios are employed to generate normal operation condition (NOC) data, as well as faults of different types and magnitudes. By simulating normal operation and faulty images in this way, it is possible to assess the monitoring potential of the proposed methodology to detect abnormal situations under a diversity of monotonically increasing faulty conditions. Results show that the proposed methodology is able to effectively detect changes in both colour and texture characteristics and one type of monitoring statistics in particular leads the performance in most of the tested scenarios: the PCA statistics for monitoring multiscale textural features. For this reason and for encompassing less computational and programming effort, its adoption is particularly recommended.

Fault Condition Recognition Based on Multi-scale Texture Features and Embedding Prior Knowledge K-means for Antimony Flotation Process

The key point to achieve automatic control and optimal operation in flotation process is to recognize the flotation conditions correctly. By the fact that it's difficult to detect and identify fault conditions in antimony flotation process, a fault condition recognition method based on multi-scale texture features extraction and k-means clustering embedding prior knowledge is proposed in this paper. Firstly, wavelet transform method is applied to froth images, and sub-images at different scales are obtained. Then neighboring gray level dependence matrix of sub-images is calculated, and the novel multi-scale texture features are statistically extracted, which can reflect the statistical laws of gray level changes in froth images. Finally, an embedding priori knowledge k-means clustering algorithm is applied to offline classification of froth images under different working conditions, and the online recognition is realized as well. Simulation and real-time recognition results show that the proposed method can achieve satisfactory performance on the fault condition recognition for antimony flotation process.

Automatic detection of the anterior and posterior commissures on MRI scans using regression forests.

Identification of the anterior and posterior commissure is crucial in stereotactic and functional neurosurgery, human brain mapping, and medical image processing. We present a learning-based algorithm to automatically and rapidly localize these landmarks using random forests regression. Given a point in the image, we extract a set of multi-scale long-range textural features, and associate a probability for this point to be the landmark. We build random forests models to learn the relationship between the value of these features and the probability of a point to be a landmark point. Three-stage coarse-to-fine models are trained for AC and PC separately using down-sampled by 4, down-sampled by 2, and the original images. Testing is performed in a hierarchical approach to first obtain a rough estimation at the coarse level and then fine-tune the landmark position. We extensively evaluate our method in a leave-one-out fashion using a large dataset of 100 T1-weighted images. We also compare our method to the state-of-art AC/PC detection methods including an atlas-based approach with six well-established nonrigid registration algorithms and a publicly available implementation of a model-based approach. Our method results in an overall error of 0.84±0.41mm for AC, 0.83±0.36mm for PC and a maximum error of 2.04mm; it performs significantly better than the model-based AC/PC detection method we compare it to and better than three of the nonrigid registration methods. It is much faster than nonrigid registration methods.

Classification of high resolution imagery based on fusion of multiscale texture features

In high resolution data classification process, combining texture features with spectral bands can effectively improve the classification accuracy. However, the window size which is difficult to choose is regarded as an important factor influencing overall classification accuracy in textural classification and current approaches to image texture analysis only depend on a single moving window which ignores different scale features of various land cover types. In this paper, we propose a new method based on the fusion of multiscale texture features to overcome these problems. The main steps in new method include the classification of fixed window size spectral/textural images from 3×3 to 15×15 and comparison of all the posterior possibility values for every pixel, as a result the biggest probability value is given to the pixel and the pixel belongs to a certain land cover type automatically. The proposed approach is tested on University of Pavia ROSIS data. The results indicate that the new method improve the classification accuracy compared to results of methods based on fixed window size textural classification.

Remote sensing of tea plantations using an SVM classifier and pattern-based accuracy assessment technique

Tea (Camellia sp.) and its plantation are very important on a worldwide scale as it is the second-most consumed beverage after water. Therefore, it becomes necessary to map the widely distributed tea plantations under various geographies and conditions. Remote-sensing techniques are effective tools to map and monitor the impact of tea plantation on land-use/land-cover (LULC). Remote sensing of tea plantations suffers from spectral mixing as these plantation areas are generally surrounded by similar types of green vegetation such as orchards and bushes. This problem is mainly tied to planting style, topography, and spectral characteristics of tea plantations, and the side effects are observed as low classification accuracies after the classification process. In this study, to overcome this problem, a three-step approach was proposed and implemented on a test area with high slope. As a first step, spectral and multi-scale textural features based on Gabor filters were extracted from high resolution multispectral digital aerial images. Similarly, based on the wavelength range of the sensor, a modified normalized difference vegetation index (MNDVI) was applied to distinguish the green vegetation cover from other LULCs. The second step involves the classification of multidimensional textural and spectral feature combinations using a support vector machine (SVM) algorithm. As a final step, two different techniques were applied for evaluating classification accuracy. The first one is a traditional site-specific accuracy assessment based on a confusion matrix calculating statistical metrics for different feature combinations. The overall accuracy and kappa values were calculated as 93.68% and 0.92, 93.82% and 0.92, and 97.40% and 0.97 for LULC maps produced by red, green, and blue (RGB), RGB + MNDVI, and RGB + MNDVI + Gabor features, respectively. The second accuracy assessment technique was the pattern-based accuracy assessment. The technique involves polygon-based fuzzy local matching. Three comparison maps showing local matching indices were obtained and used to compute the global matching index (g) for LULC maps of each feature set combination. The g values were g(RGB) (0.745), g(RGB+MNDVI) (0.745), and g(RGB+MNDVI+Gabor) (0.765) for comparison maps. Finally, based on accuracy assessment metrics, the study area was successfully classified and tea plantation features were extracted with high accuracy.

A Global Human Settlement Layer From Optical HR/VHR RS Data: Concept and First Results

A general framework for processing high and very-high resolution imagery in support of a Global Human Settlement Layer (GHSL) is presented together with a discussion on the results of the first operational test of the production workflow. The test involved the mapping of 24.3 million km <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of the Earth surface spread in four continents, corresponding to an estimated population of 1.3 billion people in 2010. The resolution of the input image data ranges from 0.5 to 10 meters, collected by a heterogeneous set of platforms including satellite SPOT (2 and 5), CBERS 2B, RapidEye (2 and 4), WorldView (1 and 2), GeoEye 1, QuickBird 2, Ikonos 2, and airborne sensors. Several imaging modes were tested including panchromatic, multispectral and pan-sharpened images. A new fully automatic image information extraction, generalization and mosaic workflow is presented that is based on multiscale textural and morphological image features extraction. New image feature compression and optimization are introduced, together with new learning and classification techniques allowing for the processing of HR/VHR image data using low-resolution thematic layers as reference. A new systematic approach for quality control and validation allowing global spatial and thematic consistency checking is proposed and applied. The quality of the results are discussed by sensor, band, resolution, and eco-regions. Critical points, lessons learned and next steps are highlighted.

BEYOND HAND-CRAFTED FEATURES IN REMOTE SENSING

Abstract. A basic problem of image classification in remote sensing is to select suitable image features. However, modern classifiers such as AdaBoost allow for feature selection driven by the training data. This capability brings up the question whether hand-crafted features are required or whether it would not be enough to extract the same quasi-exhaustive feature set for different classification problems and let the classifier choose a suitable subset for the specific image statistics of the given problem. To be able to efficiently extract a large quasi-exhaustive set of multi-scale texture and intensity features we suggest to approximate standard derivative filters via integral images. We compare our quasi-exhaustive features to several standard feature sets on four very high-resolution (VHR) aerial and satellite datasets of urban areas. We show that in combination with a boosting classifier the proposed quasi-exhaustive features outperform standard baselines.

Fast two-stage segmentation via non-local active contours in multiscale texture feature space

In this paper, a new non-local active contour model is proposed for fast unsupervised segmentation of texture images. Under our framework, problems of texture description are addressed in a texture feature space. Then, the texture features are adaptively represented across scales and their homogeneities are efficiently measured by Wasserstein metric. With total variation regularization, an external energy including a non-local term and a global term is introduced into our energy functional, which can integrate non-local patch interactions with region homogeneities inside or outside the evolving contours. Our model proportionally reaches the balance between local and global homogeneities of features and exactly extracts meaningful objects. Finally, the segmentation approach is split into two stages, coarse segmentation for fast location in the coarse-scale space and accurate segmentation for bias correction in the fine-scale space. And the two segmentation problems are reformulated into the convex optimization framework, providing a global minimizer to our active contour model. Segmentation results of the synthetic and real-world images show that our model can accurately segment object regions in a fast way.

Extraction of High Spatial Resolution Remote Sensing Image Classification based on PCA and Multi-scale Texture Feature

Extraction of High Spatial Resolution Remote Sensing Image Classification based on PCA and Multi-scale Texture Feature

Assessment of carotid atherosclerosis from B-mode ultrasound images using directional multiscale texture features

In this paper, three multiscale transforms with directional character, namely the dual-tree complex wavelet (DTCWT), the finite ridgelet (FRIT) and the fast discrete curvelet (FDCT) transforms, were comparatively assessed with respect to their ability to characterize carotid atherosclerotic plaque from B-mode ultrasound and discriminate between symptomatic and asymptomatic cases. The standard deviation and entropy of the detail subimages produced for each decomposition scheme were used as texture features. Feature selection included ranking the features according to their highest separability value and the minimum correlation among them. Due to the rather limited size of the sample population, the selected features were resampled 100 times by the bootstrap technique and divided into training and test sets. For each pair of sets, a support vector machine classifier was trained on the training set and evaluated on the test set. The average overall classification performance for systole (diastole) was 70% (65.2%), 72.6% (70.4%) and 84.9% (73.6%) for the DTCWT, FRIT and FDCT, respectively. These preliminary results showed the superiority of the curvelet transform, in terms of classification accuracy, being of great importance for the diagnosis and management of plaque instability in carotid atheromatous stenosis.

Random Forest classification of Mediterranean land cover using multi-seasonal imagery and multi-seasonal texture

A Random Forest (RF) classifier was applied to spectral as well as mono- and multi-seasonal textural features extracted from Landsat TM imagery to increase the accuracy of land cover classification over a complex Mediterranean landscape, with a large number of land cover categories and low inter-class separability. Five different types of geostatistical textural measure for three different window sizes and three different lags were applied creating a total of 972 potential input variables. Madograms, rodograms and direct variograms for the univariate case and cross- and pseudo-cross variograms for the multivariate case, together with multi-seasonal spectral information, were used in a RF classifier to map the land cover types. The pseudo-cross and cross variograms were used specifically to incorporate important seasonal/temporal information. Incorporating multi-scale textural features into the RF models led to an increase in the overall index of 10.71% and, for the most accurate classification, the increase was up to 30% in some classes. The differences in the kappa coefficient for the textural classification models were evaluated statistically using a pairwise Z-test, revealing a significant increase in per-class classification accuracy compared to GLCM-based texture measures. The pseudo-cross variogram between the visible and near-infrared bands was the most important textural features for general classification, and the multi-seasonal pseudo-cross variogram had an outstanding importance for agricultural classes. Overall, the RF classifier applied to a reduced subset of input variables composed of the most informative textural features led to the highest accuracy. Highly reliable classification results were obtained when the 16 most important textural features calculated at single scales (window sizes) were selected and used in the classification. The proposed methodology significantly increased the classification accuracy achieved with a spectral maximum likelihood classifier (ML). The kappa values for the textural RF and ML classifications were equal to 0.92 and 0.83, respectively.

Computerized Macular Pathology Diagnosis in Spectral Domain Optical Coherence Tomography Scans Based on Multiscale Texture and Shape Features

To develop an automated method to identify the normal macula and three macular pathologies (macular hole [MH], macular edema [ME], and age-related macular degeneration [AMD]) from the fovea-centered cross sections in three-dimensional (3D) spectral-domain optical coherence tomography (SD-OCT) images. A sample of SD-OCT macular scans (macular cube 200 × 200 or 512 × 128 scan protocol; Cirrus HD-OCT; Carl Zeiss Meditec, Inc., Dublin, CA) was obtained from healthy subjects and subjects with MH, ME, and/or AMD (dataset for development: 326 scans from 136 subjects [193 eyes], and dataset for testing: 131 scans from 37 subjects [58 eyes]). A fovea-centered cross-sectional slice for each of the SD-OCT images was encoded using spatially distributed multiscale texture and shape features. Three ophthalmologists labeled each fovea-centered slice independently, and the majority opinion for each pathology was used as the ground truth. Machine learning algorithms were used to identify the discriminative features automatically. Two-class support vector machine classifiers were trained to identify the presence of normal macula and each of the three pathologies separately. The area under the receiver operating characteristic curve (AUC) was calculated to assess the performance. The cross-validation AUC result on the development dataset was 0.976, 0.931, 0939, and 0.938, and the AUC result on the holdout testing set was 0.978, 0.969, 0.941, and 0.975, for identifying normal macula, MH, ME, and AMD, respectively. The proposed automated data-driven method successfully identified various macular pathologies (all AUC > 0.94). This method may effectively identify the discriminative features without relying on a potentially error-prone segmentation module.

Identification of hazelnut fields using spectral and Gabor textural features

Land cover identification and monitoring agricultural resources using remote sensing imagery are of great significance for agricultural management and subsidies. Particularly, permanent crops are important in terms of economy (mainly rural development) and environmental protection. Permanent crops (including nut orchards) are extracted with very high resolution remote sensing imagery using visual interpretation or automated systems based on mainly textural features which reflect the regular plantation pattern of their orchards, since the spectral values of the nut orchards are usually close to the spectral values of other woody vegetation due to various reasons such as spectral mixing, slope, and shade. However, when the nut orchards are planted irregularly and densely at fields with high slope, textural delineation of these orchards from other woody vegetation becomes less relevant, posing a challenge for accurate automatic detection of these orchards. This study aims to overcome this challenge using a classification system based on multi-scale textural features together with spectral values. For this purpose, Black Sea region of Turkey, the region with the biggest hazelnut production in the world and the region which suffers most from this issue, is selected and two Quickbird archive images (June 2005 and September 2008) of the region are acquired. To differentiate hazel orchards from other woodlands, in addition to the pansharpened multispectral (4-band) bands of 2005 and 2008 imagery, multi-scale Gabor features are calculated from the panchromatic band of 2008 imagery at four scales and six orientations. One supervised classification method (maximum likelihood classifier, MLC) and one unsupervised method (self-organizing map, SOM) are used for classification based on spectral values, Gabor features and their combination. Both MLC and SOM achieve the highest performance (overall classification accuracies of 95% and 92%, and Kappa values of 0.93 and 0.88, respectively) when multi temporal spectral values and Gabor features are merged. High F β values (a combined measure of producer and user accuracy) for detection of hazel orchards (0.97 for MLC and 0.94 for SOM) indicate the high quality of the classification results. When the classification is based on multi spectral values of 2008 imagery and Gabor features, similar F β values (0.95 for MLC and 0.93 for SOM) are obtained, favoring the use of one imagery for cost/benefit efficiency. One main outcome is that despite its unsupervised nature, SOM achieves a classification performance very close to the performance of MLC, for detection of hazel orchards.

Segmentation of Natural Images by Texture and Boundary Compression

We present a novel algorithm for segmentation of natural images that harnesses the principle of minimum description length (MDL). Our method is based on observations that a homogeneously textured region of a natural image can be well modeled by a Gaussian distribution and the region boundary can be effectively coded by an adaptive chain code. The optimal segmentation of an image is the one that gives the shortest coding length for encoding all textures and boundaries in the image, and is obtained via an agglomerative clustering process applied to a hierarchy of decreasing window sizes as multi-scale texture features. The optimal segmentation also provides an accurate estimate of the overall coding length and hence the true entropy of the image. We test our algorithm on the publicly available Berkeley Segmentation Dataset. It achieves state-of-the-art segmentation results compared to other existing methods.

Bayesian texture classification and retrieval based on multiscale feature vector

This paper proposes a supervised multiscale Bayesian texture classifier. The classifier exploits the dual-tree complex wavelet transform (DT-CWT) to obtain complex-valued multiscale representations of training texture samples for each texture class. The high-pass subbands of DT-CWT decomposition of a texture image are used to form a multiscale feature vector representing magnitude and phase features. For computational efficiency, the dimensionality of feature vectors is reduced using principal component analysis (PCA). The class conditional probability density function of low-dimensional feature vectors for each texture class is then estimated by using Parzen-window estimate with identical Gaussian kernels and is used to represent the texture class. A query texture image is classified as the corresponding texture class with the highest a posteriori probability according to a Bayesian inferencing. The superior performance and robustness of the proposed classifier is demonstrated for classifying texture images from image databases. The proposed multiscale texture feature vector extracted from both magnitude and phase of DT-CWT subbands of a query image is also shown to be effective for texture retrieval.

Surface grading of bamboo strips using multi-scale color texture features in eigenspace

In order to achieve high competitive quality of bamboo products, it appears that bamboo strips with naturally different tonalities should be elaborately sorted into different classes according to their global color texture appearance. Inspired by the coarse-to-fine visual perception process of human vision system, this paper proposes a new surface grading approach by integrating the color and texture of bamboo strips based on Gaussian multi-scale space. The multi-scale representations of color texture for the original image of bamboo strips could be obtained and used to construct the multivariate image, each channel of which represents a perceptual observation from different scales. The multivariate image analysis (MIA) techniques are used to extract multi-scale features from the resulting multivariate image data. The characteristic images corresponding to typical classes are selected to build the model of the reference eigenspace. The novel testing images and the training images are all projected onto this reference eigenspace to obtain their representative feature clusters. And the Bhattacharyya distance is used to estimate the similarity of the representative feature clusters between the testing images and the training images in the eigenspace. Then a k-NN classifier is adopted to classify the testing images into the given classes of training images. Comparative experiments have been carried out on a set of actual bamboo strip images and the experimental results verify the effective discrimination of multi-scale color texture eigenspace features and good classification accuracy of the proposed surface grading method.

Segmentation of multispectral high-resolution satellite imagery using log Gabor filters

Image segmentation has been recognized as a valuable approach that performs a region-based rather than a pixel-based analysis of high-resolution satellite imagery. A scheme for segmenting the multispectral IKONOS image based on frequency-domain filtering is presented. The frequency spectrum of typical landscape objects is analysed first. The spectrum curves are comparable in logarithmic coordinates rather than in Cartesian coordinates; therefore the Gabor filters are superseded by log Gabor filters to extract the multiscale texture features from panchromatic band. Edge features then are calculated from the pan-sharpened multispectral bands based on the vector field model. Finally, the texture-marked watershed segmentation algorithm is implemented and the segmentation accuracy is assessed. The experimental results show that the developed scheme generated an effective tool for automatic segmentation of multispectral high-resolution satellite imagery and suppressing the over-segmentation problem of watershed transform.