Set Of Descriptive Features Research Articles

GeoDMA—Geographic Data Mining Analyst

Remote sensing images obtained by remote sensing are a key source of data for studying large-scale geographic areas. From 2013 onwards, a new generation of land remote sensing satellites from USA, China, Brazil, India and Europe will produce in 1year as much data as 5 years of the Landsat-7 satellite. Thus, the research community needs new ways to analyze large data sets of remote sensing imagery. To address this need, this paper describes a toolbox for combing land remote sensing image analysis with data mining techniques. Data mining methods are being extensively used for statistical analysis, but up to now have had limited use in remote sensing image interpretation due to the lack of appropriate tools. The toolbox described in this paper is the Geographic Data Mining Analyst (GeoDMA). It has algorithms for segmentation, feature extraction, feature selection, classification, landscape metrics and multi-temporal methods for change detection and analysis. GeoDMA uses decision-tree strategies adapted for spatial data mining. It connects remotely sensed imagery with other geographic data types using access to local or remote database. GeoDMA has methods to assess the accuracy of simulation models, as well as tools for spatio-temporal analysis, including a visualization of time-series that helps users to find patterns in cyclic events. The software includes a new approach for analyzing spatio-temporal data based on polar coordinates transformation. This method creates a set of descriptive features that improves the classification accuracy of multi-temporal image databases. GeoDMA is tightly integrated with TerraView GIS, so its users have access to all traditional GIS features. To demonstrate GeoDMA, we show two case studies on land use and land cover change.

Analysis of parcel-based image classification methods for monitoring the activities of the Land Bank of Galicia (Spain)

The abandonment of agricultural plots entails a low economic productivity of the land and a higher vulnerability to wildfires and degradation of affected areas. In this sense, the local government of Galicia is promoting new methodologies based on high-resolution images in order to classify the territory in basic and generic land uses. This procedure will be used to control the sustainable management of plots belonging to the Land Bank. This paper presents an application study for maintaining and updating land use/land cover geospatial databases using parcel-oriented classification. The test is performed over two geographic areas of Galicia, in the northwest of Spain. In this region, forest and shrublands in mountain environments are very heterogeneous with many private unproductive plots, some of which are in a high state of abandonment. The dataset is made of high spatial resolution multispectral imagery, cadastral cartography employed to define the image objects (plots), and field samples used to define evaluation and training samples. A set of descriptive features is computed quantifying different properties of the objects, i.e. spectral, texture, structural, and geometrical. Additionally, the effect on the classification and updating processes of the historical land use as a descriptive feature is tested. Three different classification methodologies are analyzed: linear discriminant analysis, decision trees, and support vector machine. The overall accuracies of the classifications obtained are always above 90 % and support vector machine method is proved to provide the best performance. Forest and shrublands areas are especially undefined, so the discrimination between these two classes is low. The results enable to conclude that the use of automatic parcel-oriented classification techniques for updating tasks of land use/land cover geospatial databases, is effective in the areas tested, particularly when broad and well defined classes are required.

TEXTAL System: Artificial Intelligence Techniques for Automated Protein Model Building

Publisher Summary TEXTAL is a new computer program designed to build protein structures automatically from electron density maps. It uses artificial intelligence (AI) and pattern recognition techniques to try to emulate the intuitive decision making of experts in solving protein structures. TEXTAL can be a great time-saving benefit to crystallographers by automatically building accurate models for electron density maps. TEXTAL is aimed at solving maps in the 2.5–3.5 A range that are just at the range of human interpretability, but in practice, TEXTAL is aimed at solving maps constructed from multiwavelength anomalous dispersion (MAD) data. TEXTAL has the potential to reduce one of the bottlenecks of high-throughput structural genomics. As an automated model-building system, it takes an electron-density map as input and ultimately outputs a protein model (with atomic coordinates). The basic idea behind pattern recognition (at least for supervised learning) is to train the system by giving it labeled examples in several competing categories (such as tanks vs. civilian cars and trucks). Each example is represented by a set of descriptive features, which are measurements derived from the data source that characterize unique aspects of each one (for example, color and size).

NOISE-ROBUST AND INVARIANT OBJECT CLASSIFICATION BY THE HIGH-ORDER STATISTICAL PATTERN SPECTRUM

A new shape descriptor, the high order statistical pattern spectrum (HSP), able to extract from real images a set of descriptive features which can be used to classify objects regardless of their positions, sizes, orientations and the presence of noise, has been developed. The HSP is an internal, noise-robust, noninformation-preserving operator which combines the properties of invariance of the high order pattern spectrum and the properties of noise robustness of the statistical pattern spectrum. A neural network trained by a back-propagation algorithm has been used to test the method on different classification problems. Experimental results are presented on both synthetic and real images corrupted by various levels of noise and containing an object in different positions. Comparisons with other existing shape descriptor operators have been also performed.

Methodological aspects of using decision trees to characterise leiomyomatous tumors

The aim of the present work is to present the potential uses of a classification technique labeled the "decision tree" for tumor characterisation when faced with a large number of features. The decision tree technique enables multifeature logical classification rules to be produced by determining discriminatory values for each feature selected. In this report, we propose a methodology that used decision trees to compare and evaluate the information contributed by different types of features for tumor characterisation. This methodology is able to produce a set of hypotheses related to a diagnosis and or prognosis problem. For example, hypotheses can be producted (on the basis of a set of descriptive features) to explain why tumor cases belong to a given histopathological group. To illustrate our purpose, this methodology was applied to the difficult problem of leiomyomatous tumour diagnosis. The aim was to illustrate what kind of diagnostic information can be extracted from a sample data set including 23 smooth muscle tumors (14 benign leiomyomas and 9 malignant leiomyosarcomas) described by a large set of computer-assisted, microscope-generated features. Three groups of features were used relating to: (1) ploidy level determination (10 features), (2) quantitative chromatin pattern description (15 features), and (3) immunohistochemically related antigen specificities (6 features). All these features were quantified by digital cell image analysis. The results suggest that an objective distinction between leiomyomas and leiomyosarcomas can be established by means of simple logical rules depending on only a few features among which the immunohistochemically revealed antigen expression of desmin plays a preponderant part. One of the combinations of features proposed by the methodology is interesting for pathologists, because it includes two features describing the appearance of a nucleus in terms of chromatin distribution homogeneity and density, two features widely used by pathologists in tumor-grading systems.