Search

Global food production in the developing world occurs within sub-hectare fields that are difficult to identify with moderate resolution satellite imagery. Knowledge about the distribution of these fields is critical in food security programs. We developed a semi-automated image segmentation approach using wall-to-wall sub-meter imagery with high-performance computing to map crop area (CA) throughout Tigray, Ethiopia that encompasses over 41,000 km 2 . Multiple processing streams were tested to minimize mapping error while applying five unique smoothing kernels to capture differences in land surface texture associated to CA. Typically, very-small fields (mean

The characteristics of synthetic aperture radar (SAR) image texture may be related to the properties of underlying elemental scene scatterers through established models based on the properties of a backscattering coefficient (in this paper, we use and unnormalized coefficient s ) of these scatterers. In this paper, we generate raw SAR data by simulating the statistical characteristics of elemental scene scatterers such as the order of the gamma distribution, the form and length of their spatial correlation, and their spatial density. This simulation is carried out using a SAR signal simulation system called cSAR. We describe a particular set of methods to simulate image texture used in cSAR, and provide a detailed analysis of simulated s and of the speckle and texture characteristics of simulated images. We found that the distribution of s was strongly affected by the order, density, and correlation length of the underlying scatterers. We found that the simulated SAR images were consistently K -distributed as expected. The estimated image order was a strong function of the scattering properties and that the estimated image order is a relatively weak descriptor of image texture when used on its own. The correspondence between the observed image autocorrelation function (ACF) and the theoretical models of Oliver is excellent, and we could estimate the scatterer correlation length by fitting the Oliver model to the observed ACF. We combined the estimated image order and correlation length and found potential for using these two image texture descriptors in classification and segmentation algorithms.

The extraction of river planforms from remotely sensed satellite images is a task of crucial importance to many applications such as land planning, water resource monitoring or flood prediction. In this paper we present a novel framework for the extraction of rivers from Synthetic Aperture Radar (SAR) images, based on superpixel segmentation and subsequent classification. Superpixel segmentation is achieved via a modelling of the image pixels’ amplitudes and spatial coordinates as a finite mixture model, where the Generalised Gamma distribution is used to accurately model a variety of high-resolution SAR scenes. A number of features describing texture and statistics are extracted on a superpixel level, facilitating the identification of river superpixels - planforms are then extracted via unsupervised,agglomerative clustering thus eliminating the need for labelled training data. We present results of our proposed method on ICEYE-X2 and SENTINEL-1 SAR data demonstrating its ability to produce pixel-accurate river masks.

One of the main tasks of the remote sensing images classification is the formation of a system of features. We present descriptors, of image texture which consist of characteristics of random point fields formed for pixels of distinct brightness. The image is fragmented into blocks, for the elements of which the histogram is build and its local maxima are determined. The random point field is formed as a set of location of centers of every fragment with brightness extrema of the histogram. A descriptors of the image which consists of a distribution of the number of random point field elements is constructed.

In this paper, an unsupervised segmentation methodology is proposed for remotely sensed images by using Fractional Differential (FD) based texture analysis model and Iterative Self-Organizing Data Analysis Technique Algorithm (ISODATA). Essentially, image segmentation is used to assign unique class labels to different regions of an image. In this work, it is transformed into texture segmentation by signifying each class label as a unique texture class. The FD based texture analysis model is suggested for texture feature extraction from images and ISODATA is used for segmentation. The proposed methodology was first implemented on artificial target images and then on remote sensing images from Google Earth. The results of the proposed methodology are compared with those of the other texture analysis methods such as LBP (Local Binary Pattern) and NBP (Neighbors based Binary Pattern) by visual inspection as well as using classification measures derived from confusion matrix. It is justified that the proposed methodology outperforms LBP and NBP methods.

Due to the diversity of geographical objects, it makes great sense to introduce multiscale segmentation/representation into the analysis and interpretation of high-spatial-resolution remote sensing images. However, with the increasing use of high-resolution images, traditional multiscale segmentation methods gradually show their lack in efficiency, particularly when handling large-scale images. In this paper, a novel bilevel scale-sets model (BSM) is proposed for multiscale region-based representation of large-scale remote sensing images. In the BSM, first, an image is divided into blocks with overlapped margins, and a low-level scale-sets model is blockwisely implemented. Second, a segmentation result is obtained by retrieving and mosaicking the blockwise segmentation results, based on which a high-level scale-sets model is implemented covering the whole image. To further improve the efficiency of the BSM, a parallel implementation is presented for the blockwise scale-sets model. In the experiments, first, the effectiveness of the BSM is validated using a WorldView2 image covering a coastal area of Shenzhen, where the BSM obtains accurate multiscale representation results without any mosaic artifacts. Then, the efficiency of the BSM is demonstrated by comparing with the state-of-the-art multiscale segmentation method, i.e., the one integrated in the commercial software eCognition v9.2, where the proposed BSM takes about 7 min to process a 24 000 $\times$ 24 000 multispectral ZY3 image and is two to three times faster than the competing method.

The dynamic image segmentation algorithm with multiple stepwise evaluation machines was applied to resultant the new boundary from image contents. The concept of data fusion is also discussed in this research for making the good decision of image behavior by a 3D image describer. It achieves the high-understanding objects by merging some non-distinct image domains from the training patterns. Image describer contains expert knowledge to extract appropriate behaviors of the identified image patterns through the efficient dynamic image segmentation algorithm. The novel dynamic image segmentation algorithm is directly applied to explore recognitions of remote sensing images, where it can quickly choice the proper partition number of interesting image patterns area and determine their associated central positions. Due to the specific image intensity appropriately represent in the form of 3D description, an approximation object was dynamically generated with the image partition phase and merging stage to find appropriate 3D image describer. This 3D image describer explicitly presents its feature in diverse maps. Finally, the classification problems of three remote sensing images in computer simulations compared with both k-means and Fuzzy c-means (FCMs) methods. The measurement of misclassification error (ME) is selected to present the great results in various remote sensing images segmentation by the designed algorithm.

Batch-mode active learning (AL) approaches are dedicated to the training sample set selection for classification, regression, and retrieval problems, where a batch of unlabeled samples is queried at each iteration by considering both the uncertainty and diversity criteria. However, for remote sensing applications, the conventional methods do not consider the spatial coherence between the training samples, which will lead to the unnecessary cost. Based on the above two points, this paper proposes a spatial coherence-based batch-mode AL method. First, mean shift clustering is used for the diversity criterion, and thus the number of new queries can be varied in the different iterations. Second, the spatial coherence is represented by a two-level segmentation map which is used to automatically label part of the new queries. To get a stable and correct second-level segmentation map, a new merging strategy is proposed for the mean shift segmentation. The experimental results with two real remote sensing image data sets confirm the effectiveness of the proposed techniques, compared with the other state-of-the-art methods.

We present the enrichment of the Prague Texture Segmentation Data-Generator and Benchmark (PTSDB) to include the assessment of the remote sensing (RS) image segmenters. The PTSDB tool is a Web-based (http://mosaic.utia.cas.cz) service designed for real-time performance evaluation, mutual comparison, and ranking of various supervised or unsupervised static or dynamic image segmenters. PTSDB supports rapid verification and development of new segmentation approaches. The RS datasets contain ten spectral Advanced Land Imager (ALI) satellite images, their RGB subsets, and very-high-resolution GeoEye RGB images, with optional additive-noise-resistance checking. Alternative setting options allow us to also test scale, rotation, or illumination invariance. The meaningfulness of the newly proposed dataset is demonstrated by testing and comparing several RS segmentation algorithms, and showing that the benchmark figures provide a solid framework for the fair and critical comparison among different techniques.

Taking the high-resolution remote sensing image of industrial waste dump as the research object, this paper proposes a new image segmentation method based on markercontrolled watershed transform and region merging. The method gets the finial partition result by two-phrase segmentation on the pan-sharpened true color ALOS image. In the first phase, color gradient image of original image should be calculated and then it is modified by morphological impose minima, which should use markers extracted in two different ways. Lastly, preliminary segmentation result is obtained by watershed transform operates on the modified color gradient image. To solve the over-segmentation of industrial solid waste and other ground objects, region merging operation is performed according to the similarity measure criterion based on segmented objects’ color histogram Bhattacharyya coefficient in the second phrase, and then the final result is obtained. This method has been tested on the pan-sharpened ALOS image of 2.5 meters resolution in Shizuishan industrial zone, China. Experiment results demonstrate that this method is feasible and effective to segment the remote sensing industrial solid waste image.

Segmentation of real-world remote sensing images is challenging because of the large size of those data, particularly for very high resolution imagery. However, a lot of high-level remote sensing methods rely on segmentation at some point and are therefore difficult to assess at full image scale, for real remote sensing applications. In this paper, we define a new property called stability of segmentation algorithms and demonstrate that piece- or tile-wise computation of a stable segmentation algorithm can be achieved with identical results with respect to processing the whole image at once. We also derive a technique to empirically estimate the stability of a given segmentation algorithm and apply it to four different algorithms. Among those algorithms, the mean-shift algorithm is found to be quite unstable. We propose a modified version of this algorithm enforcing its stability and thus allowing for tile-wise computation with identical results. Finally, we present results of this method and discuss the various trends and applications.

Mean shift algorithm is a robust approach toward feature space analysis, which has been wildly used for natural scene image and medical image segmentation. Due to fuzzy boundary and low accuracy of Mean shift segmentation method, this paper puts forward to an improved Mean shift segmentation method of high-resolution remote sensing image based on LBP and Canny features. The results show that this improved Mean shift segmentation access can enhance segmentation accuracy compared to the traditional Mean shift.

Synthetic Aperture Radar (SAR) has become one of the important means for the ocean remote sensing detection of oil spills. The existing SAR image segmentation method has the issues of edge blur, poor contrast, non-uniform intensity image, so the segmentation effect is not ideal. This paper presents a variational level set SAR image of oil spill detection method based on fuzzy clustering. First of all, apply the threshold method on initial segmentation of the original SAR image to transform the initial segmented image as fuzzy clustering. Secondly, introduce the clustering results into the initial level set function to achieve the initial contour. Finally, add fuzzy clustering model in the level set energy function to complete the level set evolution process and get the final segmented image. This paper uses the threshold segmentation results to achieve the initialization of the variational level set function profile. In theory, it could improve the level set method for efficiency, and reduce the wrong segmentation phenomenon. The experimental results show that the SAR image segmentation method of oil spill has good segmentation qualities and is suitable for the edge complex image segmentation.

The textural information of a multitemporal set of ERS-1 and JERS-1 synthetic aperture radar (SAR) images was studied with the first- and second-order statistical measures. These measures had a higher information value for the land-cover and forest type classification than the SAR image intensity. The multitemporal approach was beneficial for the application of the textural measures; the textural parameters significantly improved the classification of land-cover and forest types. Based on the SAR image texture, the overall classification accuracy for seven land-cover types was 65%, while with the SAR image intensity, the classification accuracy was 50%, respectively. In the forest type classification based on the SAR image texture and intensity, the overall classification accuracy for four forest types was 66%, while with the intensity, the accuracy was 40%, respectively. The weather and seasonal conditions had a significant effect on the textural information of SAR images. The best separability of the signatures and the best land-cover and forest type classification accuracy was achieved under summer conditions. The snow cover and arid conditions decreased the textural information of the SAR images.

This paper proposes a novel two-stage method for remote sensing image segmentation. First, initial small segments, also called subobject primitives (sub-OPs), are obtained using edge-constrained watershed segmentation and edge allocation. These segments are gradually merged into a larger segment until the edge-controlled limits are reached, thereby creating the initial OPs. In this stage, a concept of hard-boundary ratio is proposed to control the merge effectively. Second, nonconstrained merging is conducted on the OPs, which results in final segmentation. In addition, a repeatable pairwise segment-merging scheme is utilized. This scheme improves method efficiency and accuracy. Comprehensive experiments comparing this new method with the multiresolution segmentation method of eCognition were conducted. Results show that this new method has the following advantages: 1) higher segmentation accuracy and OP boundary precision and 2) less dependence on the scale parameter.

Due to the characteristic of remote sensing image, we propose a novel method based on K-means algorithm also with the improved multi-phrase level set model. Comparing with the classical multi-phase C-V model, the improved model considers the region area information, gradient information and edge detection .Proper use of gradient information can overcome the inaccurate edge localization defects in image segmentation. The edge detection is used for keeping the boundary information better in the evolution process .For the reason of picking up the contour’s convergent speed and enable the avoidance of trap .Four stages are constructed. Firstly, a median filtering is applied to smooth the original image and reduce parts of noise. Secondly, the usage of K-means algorithm gains more obvious differences of characteristics. Next, the reconstruction of gradient is obtained by using Sobel operator. Finally, segmentation result is achieved by using an improved method of multi-phase level set image segmentation. Experimental results show that the proposed approach has advantages in rapid and efficient application of remote sensing image segmentation.

To address the problem of multi-target retrieval (MTR) of remote sensing images, this study proposes a new object-level feature representation model. The model provides an enhanced application image representation that improves the efficiency of MTR. Generating the model in our scheme includes processes, such as object-oriented image segmentation, feature parameter calculation, and symbolic image database construction. The proposed model uses the spatial representation method of the extended nine-direction lower-triangular (9DLT) matrix to combine spatial relationships among objects, and organizes the image features according to MPEG-7 standards. A similarity metric method is proposed that improves the precision of similarity retrieval. Our method provides a trade-off strategy that supports flexible matching on the target features, or the spatial relationship between the query target and the image database. We implement this retrieval framework on a dataset of remote sensing images. Experimental results show that the proposed model achieves competitive and high-retrieval precision.

This study investigates the potential of object-based texture parameters extracted from 15m spatial resolution ASTER imagery for estimating tree size diversity in a Mediterranean forested landscape in Turkey. Tree size diversity based on tree basal area was determined using the Shannon index and Gini Coefficient at the sampling plot level. Image texture parameters were calculated based on the grey level co-occurrence matrix (GLCM) for various image segmentation levels. Analyses of relationships between tree size diversity and texture parameters found that relationships between the Gini Coefficient and the GLCM values were the most statistically significant, with the highest correlation (r=0.69) being with GLCM Homogeneity values. In contrast, Shannon Index values were weakly correlated with image derived texture parameters. The results suggest that 15m resolution Aster imagery has considerable potential in estimating tree size diversity based on the Gini Coefficient for heterogeneous Mediterranean forests.

This paper proposes a remote sensing (RS) image retrieval scheme by using image scene semantic (SS) matching. The low-level image visual features (VFs) are first mapped into multilevel spatial semantics via VF extraction, object-based classification of support vector machines, spatial relationship inference, and SS modeling. Furthermore, a spatial SS matching model that involves the object area, attribution, topology, and orientation features is proposed for the implementation of the sample-scene-based image retrieval. Moreover, a prototype system that uses a coarse-to-fine retrieval scheme is implemented with high retrieval accuracy. Experimental results show that the proposed method is suitable for spatial SS modeling, particularly geographic SS modeling, and performs well in spatial scene similarity matching.

This paper discusses an approach for river mapping and flood evaluation based on multi-temporal time series analysis of satellite images utilizing pixel spectral information for image classification and region-based segmentation for extracting water-covered regions. Analysis of MODIS satellite images is applied in three stages: before flood, during flood and after flood. Water regions are extracted from the MODIS images using image classification (based on spectral information) and image segmentation (based on spatial information). Multi-temporal MODIS images from ``normal'' (non-flood) and flood time-periods are processed in two steps. In the first step, image classifiers such as Support Vector Machines (SVMs) and Artificial Neural Networks (ANNs) separate the image pixels into water and non-water groups based on their spectral features. The classified image is then segmented using spatial features of the water pixels to remove the misclassified water. From the results obtained, we evaluate the performance of the method and conclude that the use of image classification (SVM and ANN) and region-based image segmentation is an accurate and reliable approach for the extraction of water-covered regions. (c) 2012 COSPAR. Published by Elsevier Ltd. All rights reserved.

International audience; Satellite image classification is a key task used in remote sensing for the automatic interpretation of a large amount of information. Today there exist many types of classification algorithms using advanced image processing methods enhancing the classification accuracy rate. One of the best state-of-the-art methods which improves significantly the classification of complex scenes relies on Self-Dual Attribute Profiles (SDAPs). In this approach, the underlying representation of an image is the Tree of Shapes, which encodes the inclusion of connected components of the image. The SDAP computes for each pixel a vector of attributes providing a local multi-scale representation of the information and hence leading to a fine description of the local structures of the image. Instead of performing a pixel-wise classification on features extracted from the Tree of Shapes, it is proposed to directly classify its nodes. Extending a specific interactive segmentation algorithm enables it to deal with the multi-class classification problem. The method does not involve any statistical learning and it is based entirely on morphological information related to the tree. Consequently, a very simple and effective region-based classifier relying on basic attributes is presented.

Image segmentation is a traditional method in Remote Sensing and a fundamental problem in image processing applications. It has been widely used, especially with the emergence of the Geographic Object-Based Image Analysis (GEOBIA). The results of segmentation must create uniform areas, which must allow a simpler interpretation by the users and simpler representation for classification algorithms. Several algorithms were proposed through the years, using different approaches. One that is widely used in Remote Sensing applications is the Multiresolution algorithm, that is based on the region growing method. Other, which has great potential and is applied in other research areas, is available on the Image Foresting Transform (IFT) framework, which has several image operators developed primarily for medical images. The Watershed from Grayscale Marker operator uses an edge image to perform the segmentation, however, we propose an extension of the edge detection algorithm, by summing normalized gradients of each band. This work aims to evaluate and compare these two segmentation algorithms, by comparing their results through supervised egmentation from reference regions, that were defined manually by an expert user. Quality measures were evaluated by four metrics, that represent the positional adjustment based the center of gravity, intensities, size, and the amount of overlap between the segment created by the algorithms and the reference segment. We selected 21 objects of a WorldView-2 multispectral image that were used to compute the metrics. Both methods reached similar results, by comparing the aforementioned 4 metrics applied to the 21 reference regions, IFT achieved better results for majority of regions. The IFT generated segments with similar shape when compared with the references, and the multiresolution generated results with similar sizes and positional adjustments. It may be explained by the fact that IFT uses an edge image to perform the segmentation. Both algorithms obtained similar agreement for intensity.

According to the characteristics of high-resolution remote sensing (RS) images, a new multifeature segmentation method of high-resolution remote sensing images combining the spectrum, shape, and texture features based on graph theory is presented in the paper. Firstly, the quadtree segmentation method is used to partition the original image. Secondly, the spectrum, shape, and texture weight components are calculated all based on the constructed graph. The matching degree between pixels and the texture is computed similarity. Finally, the ratio cut standards combination of the spectrum, shape, and texture weight components is used for the final segmentation. The experimental results show that this method can obtain more ideal results and higher segmentation accuracy applied to RS image than those traditional methods.

International audience; A new procedure is proposed for agricultural land-use mapping that addresses a known weakness of classical per-pixel methods in situations involving mixed tree crops. The proposed scheme uses a pair of very-high resolution satellite-borne panchromatic and multispectral images and integrates classification results of two parallel and independent analyses, respectively based on spectral and textural information. The multispectral image is divided into spectrally homogeneous but non-contiguous segments using unsupervised classification. In parallel, the panchromatic image is split into a grid of square blocks on which is performed a texture-driven supervised classification. Finally, the spectral and the textural classifications are fused to generate the land-use map. This method contrasts with object-based methods that sequentially perform image segmentation and classification. Results are evaluated both quantitatively and qualitatively, based on field survey ground-truth data. The quantitative assessment is presented in terms of overall accuracy (from 80% to 100% depending on the area) and Kappa coefficients. Visual comparison of the resulting map with the ground-truth is performed, with the analysis of the binary error maps. Merging spectral and textural classifications results in finer border delimitation and improves the overall classification accuracy of agricultural land-use by 27% as compared to textural classification alone.

This paper presents a novel method for extracting road from high resolution remote sensing image based on Gaussian mixture model and Markov random field (GMM-MRF) model optimized by the iterated conditional model (ICM). In this paper, we first divide remote sensing image into 3 classes by GMM-MRF texture segmentation model, and then extract road by fuzzy connectedness. Finally, we use shape features and mathematical morphology to do the post-processing to remove false alarm area. Since fuzzy connectedness lacks regional structure factors, it can not accurately describe the pixel similarity. By introducing image texture feature to the concept of fuzzy connectedness, we can establish texture fuzzy connectedness concept to make it more accurate description of the degree of association among pixels, which reflects the characteristics of the local and regional pixel similarity. Besides, our method choose only one road seed point and can achieve better road extraction effectiveness than fuzzy connectedness without using segmentation algorithms, which usually needs more than 15 seed points. Experiments show that the proposed method can effectively eliminate the non-road areas that whose characteristics are similar to road, and it can obtain better performance for road extraction from high resolution remote sensing image compared with fuzzy c-means and mathematical morphology.

In this paper, we proposed a new classification method based on support vector machine (SVM) combined with multi-scale segmentation. The proposed method obtains satisfactory segmentation results which are based on both the spectral characteristics and the shape parameters of segments. SVM method is used to label all these regions after multiscale segmentation. It can effectively improve the classification results. Firstly, the homogeneity of the object spectra, texture and shape are calculated from the input image. Secondly, multi-scale segmentation method is applied to the RS image. Combining graph theory based optimization with the multi-scale image segmentations, the resulting segments are merged regarding the heterogeneity criteria. Finally, based on the segmentation result, the model of SVM combined with spectrum texture classification is constructed and applied. The results show that the proposed method can effectively improve the remote sensing image classification accuracy and classification efficiency.

Because of the particularity of high resolution remote sensing image such as Quickbird, the segmentation methods used in common video or image are unsuitable for high-resolution remote sensing image. In order to overcome the shortcomings of over-segmentation existing in watershed algorithm, we bring in a new method which is an improvement to the marker-based watershed algorithm. Firstly, the median filtering is carried out to the high-resolution remote sensing image to eliminate random noise in the image, and then gradients are obtained with mathematical morphology; finally, the maker-based watershed algorithm is applied to the gradients to segment the image. Segmentation experiments carried out for the Quickbird images with different kinds of land cover types, show that the method is efficient and it can avoid over-segmentation when the image texture is homogeneous, but it also has some limitations when the image texture is relatively complex.

Information about forest biomass distribution is important for sustainable forest management and monitoring fuelwood supply. The objective of this study is to develop an accurate forest biomass map for Kampong Thom Province, Cambodia. We used a new technique (object-based approach) and a conventional technique (pixel-based approach) for the estimation of forest biomass using Landsat Enhanced Thematic Mapper Plus (ETM+). The object-based approach created segments of images, and calculated statistical and textural attributes. Our results showed that estimation accuracy of the object-based approach, with the use of band 1 and an exponential fit, was the best (R2 = 0.76), and this accuracy was comparable to that of the pixel-based approach (R2 = 0.67). Although several textural variables were related to forest biomass, they did not contribute significantly to improvement of estimation accuracy. However, the object-based method can be used for image segmentation so that the image objects are spectrally more homogeneous within individual regions than with their neighbors. Hence, they can be regarded as management units for policy-related spatial decisions. Therefore, it is possible to select either of the two methods depending upon what the situation demands.

In this paper, we propose a new algorithm for the segmentation of multiresolution remote-sensing images, which fits into the general split-and-merge paradigm. The splitting phase singles out clusters of connected regions that share the same spatial and spectral characteristics. These clusters are then regarded as atomic elements of more complex structures, particularly textures, that are gradually retrieved during the merging phase. The whole process is based on a recently developed hierarchical model of the image, which accurately describes its textural properties. In order to reduce the computational burden and preserve contours at the highest spatial definition, the algorithm works on the high-resolution panchromatic data first, using low-resolution full spectral information only at a later stage to refine the segmentation. It is completely unsupervised, with just a few parameters set at the beginning, and its final product is not a single segmentation map but rather a sequence of nested maps which provide a hierarchical description of the image, at various scales of observations. The first experimental results, obtained on a remote-sensing Ikonos image, are very encouraging and confirm the algorithm potential.

Cataloged from PDF version of article. The object-based analysis of remotely sensed imagery provides valuable spatial and structural information that is complementary to pixel-based spectral information in classi- fication. In this paper, we present novel methods for automatic object detection in high-resolution images by combining spectral information with structural information exploited by using image segmentation. The proposed segmentation algorithm uses morphological operations applied to individual spectral bands using structuring elements in increasing sizes. These operations produce a set of connected components forming a hierarchy of segments for each band. A generic algorithm is designed to select meaningful segments that maximize a measure consisting of spectral homogeneity and neighborhood connectivity. Given the observation that different structures appear more clearly at different scales in different spectral bands, we describe a new algorithm for unsupervised grouping of candidate segments belonging to multiple hierarchical segmentations to find coherent sets of segments that correspond to actual objects. The segments are modeled by using their spectral and textural content, and the grouping problem is solved by using the probabilistic latent semantic analysis algorithm that builds object models by learning the object-conditional probability distributions. The automatic labeling of a segment is done by computing the similarity of its feature distribution to the distribution of the learned object models using the Kullback–Leibler divergence. The performances of the unsupervised segmentation and object detection algorithms are evaluated qualitatively and quantitatively using three different data sets with comparative experiments, and the results show that the proposed methods are able to automatically detect, group, and label segments belonging to the same object classes.

High resolution remote sensing imagery can provide more useful information, such as spectral, shape and texture information. However, traditional pixel-based image classification approaches may suffer the increase of within-class spectral variation with improved spatial resolution. This paper presents a novel method which combines the optimal segmentation scale with Bag-of-Visual Words (BOV) representation for object-oriented classification. More precisely, an improved estimation of scale parameter (ESP) tool is adopted to determine the optimal parameters in multi-scale image segmentation. BOV is introduced to construct the midlevel representations instead of low-level features for object description. Then Support vector machine (SVM) is used for classification. And the experiments are conducted on high spatial resolution images to validate the proposed algorithm.

In this article, a newsegmentation method based on traditional region growing (RG) is proposed for high resolutionremote sensing imagery. This method takes regional minima from horizontal and vertical gradient maps of the image as seeds for the following region growing processing. The new method consists of several steps as follows: (1)deriving a morphological gradient map from the input multispectral image, (2) morphologically filtering the gradient image to remove local minima with small depthand extracting regional minima of flat areas in the resulting filtered image as seeds, (3) segmenting the multispectral image using the RG approach with reference to the seeds, and (4) merge the resulting initial segments to yield asegmentation map. In a test with a WorldView-2 multispectral image, the proposed method offered segmentation maps with nearly the same accuracy as several current methods.

Texture segmentation is a critical step in building-based analysis in urban remote sensing images to obtain more detail information for further applications. Most existing segmentation algorithms rely on region or edge information to segment, which failed to segment building surfaces with almost same texture and unclear edge between the surfaces. Therefore, in order to solve this challenging task, based on the ORTSEG algorithm in Michael T. McCann's paper, an improved ORTSEG algorithm is proposed, in which, a sparse non-negative matrix factorization (SNMF) is used for the optimization. The final segmentation results show the superiority of this improved ORTSEG algorithm.

In this paper a new object-oriented segmentation method for high-resolution remote sensing images is proposed. To limit computational complexity, a preliminary superpixel representation of the image is obtained by means of a suitable watershed transform. Then, a region adjacency graph is associated with the superpixels, with edge weights accounting for region similarity/dissimilarity. The final segmentation is then obtained by means of a graph-cutting approach, following a correlation clustering formulation. The optimal cut can be obtained by solving a Integer Linear Programming (ILP) problem, whose complexity, however, grows rapidly with the image size. Much faster near-optimal solutions are obtained, here, with a greedy solution. Experiments on a real-world high-resolution remote sensing image prove the potential of the approach.

As the application of high resolution remote sensing image is more and more popular, traditional pixel-based classification method reduces the classification accuracy, wastes a large amount of redundant spatial data and generates image which contains noise classification results. This paper focuses on the object-oriented technology research of high-resolution remote sensing image classification. Its core processing steps include image segmentation and image classification.

This paper presents a map-guided sea ice classification system built to work in parallel with the Canadian Ice Service (CIS) operations to produce pixel-based ice maps that complement actual "egg code" maps produced by CIS. The system uses the CIS maps as input to guide classification by providing information on the number of ice types and their final label for specific regions. Segmentation is based on a modified adaptive Markov random field (MRF) model that uses synthetic aperture radar (SAR) intensities and texture features as input. The ice type labeling is performed automatically by gathering evidences based on a priori information on one or two classes and deducing the other labels iteratively by comparing distributions of segments. Three methods for comparing the segment distributions (Fisher criterion, Mahalanobis distance, and Kolmogorov-Smirnov test) were implemented and compared. The system is fully described with special attention to the labeling procedure. Examples are presented in the form of two CIS SAR-based ice maps from the Gulf of Saint Lawrence region and one example from the Beaufort Sea. The results indicate that when the segmentation is good, the labeling attains best results (between 71% and 89%) based on evaluation by a sea ice analyst. Some problems remain to be assessed which are primarily attributable to discrepancies in the information provided by the egg code and what is actually visible in the SAR image. Subscale information on floe size and shape available to human analysts, but not in this classification system, also appear to be a critical information for separating some ice types.

In this study, a segmentation procedure is proposed, based on grey-level and multivariate texture to extract spatial objects from an image scene. Object uncertainty was quantified to identify transitions zones of objects with indeterminate boundaries. The Local Binary Pattern (LBP) operator, modelling texture, was integrated into a hierarchical splitting segmentation to identify homogeneous texture regions in an image. We proposed a multivariate extension of the standard univariate LBP operator to describe colour texture. The paper is illustrated with two case studies. The first considers an image with a composite of texture regions. The two LBP operators provided good segmentation results on both grey-scale and colour textures, depicted by accuracy values of 96% and 98%, respectively. The second case study involved segmentation of coastal land cover objects from a multi-spectral Compact Airborne Spectral Imager (CASI) image, of a coastal area in the UK. Segmentation based on the univariate LBP measure provided unsatisfactory segmentation results from a single CASI band (70% accuracy). A multivariate LBP-based segmentation of three CASI bands improved segmentation results considerably (77% accuracy). Uncertainty values for object building blocks provided valuable information for identification of object transition zones. We conclude that the (multivariate) LBP texture model in combination with a hierarchical splitting segmentation framework is suitable for identifying objects and for quantifying their uncertainty.

Remote sensing image segmentation is the basis of object-oriented classification of remote sensing images. It is important for the application of remote sensing images. High-resolution remote sensing images contain rich spatial texture information. SRM is an efficient image segmentation algorithm. This paper presents a segmentation algorithm to take full advantage of the high-resolution remote sensing image texture information based on LBP and SRM, in the process of merging, according to the characteristics of regions, select the appropriate method to merge. It works well in the segmentation of high-resolution remote sensing images.

The aim of this work was to develop a system based on multifeature texture analysis and modular neural networks that will facilitate the automated interpretation of satellite cloud images. Such a system will provide a standardized and efficient way for classifying cloud types that can he used as an operational tool in weather analysis. A series of 98 infrared satellite images from the geostationary satellite METEOSAT7 were employed, and 366 cloud segments were labeled into six cloud types after combined agreed observations from ground and satellite. From the segmented cloud images, nine different texture feature sets (a total of 55 features) were extracted, using the following algorithms: statistical features, spatial gray-level dependence matrices, gray-level difference statistics, neighborhood gray tone difference matrix, statistical feature matrix, Laws' texture energy measures, fractals, and Fourier power spectrum. The neural network self-organizing feature map (SOFM) classifier and the statistical K-nearest neighbor (KNN) classifier were used for the classification of the cloud images. Furthermore, the classification results of the nine different feature sets were combined, improving the classification yield for the six classes, for the SOFM classifier to 61 % and for the KNN classifier to 64%. 41 11 PART I 2662 2668 Cited By :56

Image segmentation is an essential step toward higher level image processing in remote sensing. However, the traditional image segmentation approaches based on pixels spectral characteristics and single-scale image information extraction methods have obvious flaws in this respect. Currently, multi-scale image segmentation is seen as a promising alternative of traditional segmentation method and is one of the most useful approaches in object oriented classification of remotely sensed images. In this paper, we present a multi-scale segmentation method based on Minimum Heterogeneity Rule (MHR) for merging objects. Segmentation results show that this method can easily adapt its scale parameter to different scale image analysis tasks and any chosen scale object-extraction of interest.

In this paper, the latest watershed segmentation process is actually found in the object feature extraction process. In the proposed method, initially the actual visual features are usually taken from the images using the spatial spectral heterogeneity method. Afterwards, the object features are usually taken from the new watershed segmentation method in which segmented objects are usually grouped with the PSO-SVM method. With PSO-SVM, the actual SVM parameters are usually optimized to achieve higher classification accuracy. Then similar scene images from the data base are usually taken from the SS modelling. A further variety of remote sensing images are utilized in the overall performance analysis process. The particular implementation benefits show the effectiveness of proposed new watershed segmentation method in RSIR and the reached advancement in sensitivity and also recall measures. Moreover, the actual overall performance of the proposed technique is actually considered by comparing with all the existing RSIR and the typical SBRSIR methods.

Development associated with oil and gas exploration has expanded rapidly in Alberta and Northwest Territories, Canada. Such explorations result in landscape disturbances including forest cuts, seismic lines, well and waste sites. This paper describes a novel methodology for automatic extraction of well sites from Landsat-5 TM imagery. The method combines pixel-based and object-based image analyses and contains three major steps: geometric enhancement, segmentation, and well site extraction. For accuracy assessment, a small part of the image was used and the results were compared against visual counting of well sites visible in the pan-sharpened image of Landsat-8 of the same area. Results show correctness, completeness and quality factors of 87.3%, 96.2%, and 83.7%, respectively

High spatial resolution remote sensing images provide many rich features, such as spectrum, shape, texture,etc. However, only spectral character is adopted in many traditional image segmentation methods, leading to segmentation results unsatisfactory. A multi-feature and multi-band region segmentation algorithm(MM-RSA) is proposed. First, texture image of a band is extracted and is combined into multi-spectral image. Second, seed region is selected from the combined multi-spectral image using Fuzzy C-Means Clustering method. Third, the segmentation process is performed by employing a region growing criterion, which integrates spectral and shape feature information. The algorithm not only integrates the criterions of spectrum, texture and shape, but also is of multi-scale characteristic. Experiments were conducted on a QuickBird image to evaluate the performance, and the results showed that the MM-RSA is able to effectively obtain segmentation results at different scales, and the overall performance of segmentation is improved when compared with pixel-based segmentation algorithm and multi-resolution-based segmentation algorithm.

Remote sensing technology is widely used in land cover survey. In this experiment, firstly VSVI (vegetation sample-based vegetation index) method is used to extract the forests in experiment area, and then adding texture feature to distinguish the natural forest and plantation in forests class. Then average contrast of objects (ACO) method is used to determine the best scale of chessboard segmentation when classify, and classification accuracy of different scales is compared. It is proved that the optimal segmentation scale through ACO and the best classification results have certain consistency, and achieved high classification accuracy.

Image registration is central problem to many tasks in digital image processing and therefore it has a vast range of applications. A scheme of automatic image registration based on plain objects detection and recognition in remote sensing tasks is presented in this paper. It relates to the concept of plain object in image and includes three subsystems. A subsystem of plain objects detection in image is based on a new method of plain objects detection in image. It relates to the land use classification process, different image segmentation methods and depends on the particular application. A subsystem of conjugate plain objects determination relate to a new method of plain objects recognition in images by shape based on stochastic geometry. It is invariant with respect to projective distortions. A subsystem of image registration determines an optimal geometric transformation and register given input images based on it. This paper reviews existing methods of image registration and emphasizes parametric, non-parametric and hybrid image registration techniques. Experimental results confirm the efficiency of the proposed system.

This paper applies the concepts of optimized texture segmentation to the classification of SAREX-92 data from the Amazon rain forest. Initially, a simple scene is classified using both SAR texture and Band 5 Landsat TM imagery, yielding forest and not-forest joint probabilities of 97.8% and 96.5%, respectively. When the same procedure is applied to a more complicated scene, including regenerating areas, the equivalent results are 93.8% and 67.3%, When predictable corrections for shadowing and the presence of a highway are introduced, the not-forest joint probability is improved to about 78%. The residual discrepancy is then a consequence of the different ways in which the SAR texture and TM intensity respond to regenerating areas in the scene.

It is shown empirically that information in an image of sea ice exists in two forms-textural and structural. It is shown that such information can be extracted from the conditional probability of intensity differences between neighboring pixels. There exists a relationship between the two forms of information, called the information curve, which has identifiable subsections, each related to an ice type. The overall state of the sea ice image, composed as it is of a mixture of ice types, can therefore be given by the location of phase transition points that mark the end points of the subsections on the information curve composed of the two types of information. This result leads to a segmentation of the image by ice type. The properties of the information within the image are examined. It is shown that the textural information associated with individual ice pieces is scale invariant, whereas the structural information associated with the size and shape of these pieces, which dominates the image, has a scaling or fractal property. These results led to the examination of similarity sets, describing scaled intensity differences and separation jointly. Significantly, a multifractal relationship for these sets holds at all intensities, at least for ice types associated with brittle fracture. Another result inspired by the need to identify paths of least resistance through an ice field is the existence of a distinct change in the connectivity of these similarity sets at a critical value of the index parameter of the set. Such a critical value is in turn related to that of the local textural information, which is related to the phase transitions associated with the relationship between textural and structural information.

In this paper, we tackle the problem of estimating textural parameters. We do not consider the problem of texture synthesis, but the problem of extracting textural features for tasks such as image segmentation. We take into account nonstationarities occurring in the local mean. We focus on Gaussian Markov random fields for which two estimation methods are proposed, and applied in a nonstationary framework. The first one consists of extracting conditional probabilities and performing a least square approximation. This method is applied to a nonstationary framework, dealing with the piecewise constant local mean. This framework is adapted to practical tasks when discriminating several textures on a single image. The blurring effect affecting edges between two different textures is thus reduced. The second proposed method is based on renormalization theory. Statistics involved only concern variances of Gaussian laws, leading to Cramer-Rao estimators. This method is thus especially robust with respect to the size of sampling. Moreover, nonstationarities of the local mean do not affect results. We then demonstrate that the estimated parameters allow texture discrimination for remote sensing data. The first proposed estimation method is applied to extract urban areas from SPOT images. Since discontinuities of the local mean are taken into account, we obtain an accurate urban areas delineation. Finally, we apply the renormalization based on method to segment ice in polar regions from AVHRR data.

Watershed transformation is a powerful image segmentation tool recently developed in mathematical morphology. In order to segment images initially oversegmented by watershed transformation, two approaches are considered: one is the thresholding of the gradient image proposed by us which is capable of keeping more salient image contours; the other is the well known centroid linkage region growing algorithm which merges regions with certain statistical similarities. By choosing suitable thresholds in the two approaches, hierarchical image segmentation algorithms can be constructed. A Ratio of Averages (ROA) edge detector is proposed to replace the morphological edge detectors prior to watershed transformation when applied to Synthetic Aperture Radar (SAR) images. Applications to SAR agricultural image segmentation with these hierarchical segmentation algorithms are presented. It is demonstrated that the algorithms are efficient in the segmentation of the SARimages and appropriate for land use applications w...