Search

Many organisms rely on reedbed habitats for their existence, yet, over the past century there has been a drastic reduction in the area and quality of reedbeds in the UK due to intensified human activities. In order to develop management plans for conserving and expanding this threatened habitat, accurate up-to-date information is needed concerning its current distribution and status. This information is difficult to collect using field surveys because reedbeds exist as small patches that are sparsely distributed across landscapes. Hence, this study was undertaken to develop a methodology for accurately mapping reedbeds using very high resolution QuickBird satellite imagery. The objectives were to determine the optimum combination of textural and spectral measures for mapping reedbeds; to investigate the effect of the spatial resolution of the input data upon classification accuracy; to determine whether the maximum likelihood classifier (MLC) or artificial neural network (ANN) analysis produced the most accurate classification; and to investigate the potential of refining the reedbed classification using slope suitability filters produced from digital terrain data. The results indicate an increase in the accuracy of reedbed delineations when grey-level co-occurrence textural measures were combined with the spectral bands. The most effective combination of texture measures were entropy and angular second moment. Optimal reedbed and overall classification accuracies were achieved using a combination of pansharpened multispectral and texture images that had been spatially degraded from 0.6 to 4.8 m. Using the 4.8 m data set, the MLC produced higher classification accuracy for reedbeds than the ANN analysis. The application of slope suitability filters increased the classification accuracy of reedbeds from 71% to 79%. Hence, this study has demonstrated that it is possible to use high resolution multispectral satellite imagery to derive accurate maps of reedbeds through appropriate analysis of image texture, judicious selection of input bands, spatial resolution and classification algorithm and post-classification refinement using terrain data. [ABSTRACT FROM PUBLISHER]

Airborne sensor image texture derived following a geostatistical analysis can increase the accuracy of forest classification because the resulting texture is insensitive to random variations in spectral response but related to the structural features of interest at the scale of a forest inventory (e.g. tree species). The combination of spectral and textural data derived from a kriging surface provided 86% classification accuracy in 36 pure and mixed-wood stands in seven forest classes in Alberta. This is an increase over the classification accuracy obtained when texture was derived from the original image data, and when the spectral response patterns were used alone. [ABSTRACT FROM AUTHOR]

Human colour vision has the ability to recover a measure of contextdependent surface reflectance from observed areas. This can be seen as an analogue to radiometric corrections employed in remote sensing. Procedures based on human colour vision applied in the context of remote sensing could simplify image preprocessing and classification. This study evaluates an algorithm based on Land's colour vision or 'retinex' theory. When tested on remotely sensed images, the algorithm had difficulty coping with the presence of image texture. A new framework is presented that adapts Land's procedure to processing of remotely sensed images using adaptive thresholding techniques. [ABSTRACT FROM AUTHOR]

Although research with digital airborne remote sensing data has been undertaken in different ecoregions to classify forested areas, the potential role of such imagery in deriving information to assist forest management has not yet been fully defined. The objective of this study was to determine the extent that the addition of texture could improve spectral classification of high spatial resolution images (pixel size 1m). These images represented pure and mixed wood forest stands from ecoregions in Alberta and New Brunswick, Canada. This study employed a judicious, selective application of texture to stands within a hierarchical classification framework. In Alberta, the addition of texture made a modest improvement in classification accuracy from 60% to 65%. In New Brunswick, the application of texture to selected land cover types resulted in an overall 12% improvement in classification accuracy. The addition of image texture increased classification accuracy for high spatial detail imagery relative to low spatial detail imagery. Incorporating texture into classification also improved classification accuracies for hardwood stands more so than for softwood stands, but greater attention to stand structure and composition will be needed in future work. Classification accuracies on the order of 60-65% were achieved with simple texture derivatives, maximum likelihood decision rules and conventional classification methods. [ABSTRACT FROM AUTHOR]

This study used four near-real-time multispectral Systeme pour l'Observation de la Terre (SPOT) high-resolution visible (HRV ) images to establish land cover and forest classes of relevance to slash-and-burn agriculture. The study was conducted in 1.43 million ha of the Alternative to Slash-and-Burn (ASB) global benchmark research area for Africa selected as representative of the entire Congolese basin of Central Africa and located in Southern Cameroon. The land cover and forest classes mapped have different impacts on global phenomena as well as their own management challenges and include slash-and-burn dominant farmlands, C hromolenea orodata dominant short-fallows, Imperata cylindrica dominant weeds, long-fallows or regenerated forests, raphia palm-dominant lowlands, permanently flooded swamp forests, and primary and secondary forests. In order to map these distinct and complex classes the study proposes and implements a piecemeal approach to classification involving stratification of the image into several distinct discrete subsets of forest corridors, lowlands, and uplands. The approach involves the use of image texture indicators in conjunction with groundtruth data to divide the image into discrete subsets, performing unsupervised classification on these discrete subsets, masking problem classes and reclassifying them, editing certain spectrally inseparable areas, adopting post-classification strategies, and finally mosaicking the discrete subsets into one seamless image. This approach led to an overall mapping accuracy of 82.4% with individual classes mapped at accuracies above 72.9% (user's), and above 66.7% (producer's). [ABSTRACT FROM AUTHOR]

A study of textural changes is presented concerning the nematic phase of 4- n-alkyloxybenzoic acids (in particular, 4-n-heptyl- and 4-n-octyl-oxybenzoic acids), by means of a statistical approach to the image data observed by polarized light microscopy (orthoscopic mode). A new image processing method is developed in order to detect with high sensitivity any structural change in the image frame. To do this, a set of parameters is introduced, charaterizing the observed textures. Such a set is a vector, working like a pathfinder strongly increasing the human eye or in general the sensor skilfulness to appreciate any change of the optical texture, both in space and in time. This is suitable for revealing smooth transitions, such as phase transitions between smectic and nematic phases (or between different smectic phases), or order transitions, like alignment transitions in poorly oriented nematic layers. In fact, by using this method for detecting the order transition between two nematic 'subphases' of 4-n-alkyloxybenzoic acids, the sensitivity turns out to be enhanced by a factor higher than 10 with respect to that for standard techniques. The new method allows us to define a characteristic size of the image texture: this concept is applied to analyse several image data for estimating the mean size of the domains appearing in the smectic and in the nematic phase of the compounds under study. [ABSTRACT FROM AUTHOR]

Land cover classification is a vital application area in the satellite image processing domain. Texture is a useful feature in land cover classification. In this paper, we propose a distributed texture-based land cover classification algorithm using Hidden Markov Model (HMM). Here, HMM is used for texture-based classification of remotely sensed images. Furthermore, to enhance the performance, data-intensive remotely sensed image is segmented and distributed into parallel sessions. Experiments were conducted on IRS P6 LISS-IV data, and the results were evaluated based on the confusion matrix, classification accuracy, and Kappa statistics. These results indicate that the proposed algorithm achieves a classification accuracy of 88.75%. [ABSTRACT FROM AUTHOR]

Utilization of online medical images was limited due to the lack of effective search methods, and text-based searches have been a dominating approach for the medical database management. Medical images of various modalities cannot be effectively indexed or organized with traditional text-based retrieval techniques. This has led to the use of image content for the processing and organizing the database, constituting the so-called content-based image retrieval systems. This paper presents two proposed methods, namely fuzzy connectedness image segmentation with geometric moments (FCISGMs), and localized entropy-based medical image retrieval (LEBIR) for retrieval of Digital Imaging and Communications in Medicine images. FCISGM exploits shape features for precise image retrieval by using fuzzy connectedness image segmentation. LEBIR uses localized entropy for minimizing number of computation which results in efficient image retrieval. Experimental evaluation reveals that the proposed methods outperform the existing methods in terms of precision and recall. [ABSTRACT FROM PUBLISHER]

Habitat-type land cover on Horn Island, Mississippi, northern Gulf of Mexico, was estimated for the years 1940 and 2010 using a combination of panchromatic imagery and 2010 ground survey data. A grey-level co-occurrence matrix was applied to compute reflectance coefficient of variation (CV) and texture indices. The relationships of 2010 CV ranges with known habitat types defined training regions of interest in the 1940 imagery as a substitute for 1940 ground data. Texture indices contrast, correlation, energy and entropy then served as input bands for maximum likelihood classifications which produced 1940 and 2010 habitat maps. Analysis determined that wetter habitats on Horn expanded linearly over the seven-decade period. This is attributed to constraints on sediment supply and the impacts of severe storms which led to decreases in soil depth to the water table. If this trend continues, marsh habitat will cover 31% of Horn Island’s land area by 2050. [ABSTRACT FROM PUBLISHER]

This article presents the results of discrimination of 11 wheat grain varieties. The statistical analysis included reduction of variables to a set of 49 textures with the highest discriminating strength and multidimensional analysis. Reduction of variables was performed by the following methods: genetic algorithms (sequential forward floating search [SFFS] method) as well as theClass RankerandClass RankersSearchmethods. Furthermore, the multidimensional analysis was performed by methods employing the following classifiers:Bayes, Lazy, Meta, Decision trees,andDiscriminatory analyses.The classification of individual varieties, regardless of the year of cultivation, was between 98 and 100%. [ABSTRACT FROM AUTHOR]

Airborne light detection and range (LiDAR) with three-dimensional product acquisition capabilities has become an important research hot spot. The classification and information extraction of LiDAR were usually carried out with other earth observation (EO) data (VIR/SAR, spaceborne/airborne). However, if these EO data are not always available, a particularly interesting possibility is to use only the products of LiDAR including the digital elevation model, the digital surface model and the intensity image. In this article, at the study site of Beauport, around Lac-Saint-Charles, Québec, Canada, only airborne LiDAR products are applied to realise land cover interpretation and classification. A two-hierarchy decision-tree classification is suggested and carried out. Six land cover classes are extracted, including water bodies, bare soils, impervious areas, vegetation, buildings and high-voltage electric wires. All the reference and checking data are based on 1:20,000 topographic map of Québec. According to the accuracy statistics with 966 independent checking data, it is an efficient attempt in land cover classification using only airborne LiDAR products in this site. The feasibility of the methods and the classification data sources are verified for use in the scale of 1:25,000–50,000. [ABSTRACT FROM AUTHOR]