Structural Feature Extraction Research Articles

A Fractional-order Sobel Operator for Medical Image Structure Feature Extraction

This paper proposes a novel fractional-order gradient operator for medical image structure feature extraction. The proposed operator can be seen as generalization of the first-order Sobel operator. The goal is to utilize the frequency characteristic of the fractional derivative for extracting more structure feature details. A thresholding is set based on the average fractional-order gradient for marking the edge points, and then the image structure can be extracted. Experiments show that the proposed fractional-order operator yields good visual effects.

Generating Chinese characters based on stroke splitting and feature extraction

There is a large demand for more fashionable style Chinese characters in advertising, art designing and publishing markets. However, it becomes challenging to create a new font style for so many Chinese characters (over 10,000). To solve this problem, a comprehensive Chinese fonts generating scheme is proposed in this paper. Firstly, a decomposition database for stroke splitting and feature extraction is proposed. Secondly, stroke segmentation rules are defined based on splitting, merging and structural model, location definition and minimum feature extraction. Thirdly, a radical searching algorithm based on stroke splitting is presented. Finally, it is realized that the generated characters can be zoomed, rotated and moved. Experimental result shows that Chinese characters with a new style can be generated rapidly with the proposed scheme. The created characters fit the real ones well with a high fidelity of 96.4%. The usability tests are run and participants’ subjective report show that the performance from the generated characters is similar to the original characters in both recognizability test and style-consistency test. The fonts generating method is also reliable for the other stroke constructed block characters such as Japanese and Korean characters.

Exhaustively characterizing feasible logic models of a signaling network using Answer Set Programming.

Motivation: Logic modeling is a useful tool to study signal transduction across multiple pathways. Logic models can be generated by training a network containing the prior knowledge to phospho-proteomics data. The training can be performed using stochastic optimization procedures, but these are unable to guarantee a global optima or to report the complete family of feasible models. This, however, is essential to provide precise insight in the mechanisms underlaying signal transduction and generate reliable predictions.Results: We propose the use of Answer Set Programming to explore exhaustively the space of feasible logic models. Toward this end, we have developed caspo, an open-source Python package that provides a powerful platform to learn and characterize logic models by leveraging the rich modeling language and solving technologies of Answer Set Programming. We illustrate the usefulness of caspo by revisiting a model of pro-growth and inflammatory pathways in liver cells. We show that, if experimental error is taken into account, there are thousands (11 700) of models compatible with the data. Despite the large number, we can extract structural features from the models, such as links that are always (or never) present or modules that appear in a mutual exclusive fashion. To further characterize this family of models, we investigate the input–output behavior of the models. We find 91 behaviors across the 11 700 models and we suggest new experiments to discriminate among them. Our results underscore the importance of characterizing in a global and exhaustive manner the family of feasible models, with important implications for experimental design.Availability: caspo is freely available for download (license GPLv3) and as a web service at http://caspo.genouest.org/.Supplementary information: Supplementary materials are available at Bioinformatics online.Contact: santiago.videla@irisa.fr

Discriminating bioluminescent proteins by incorporating average chemical shift and evolutionary information into the general form of Chou's pseudo amino acid composition

Bioluminescent proteins are highly sensitive optical reporters for imaging in live animals; they have been extensively used in analytical applications in intracellular monitoring, genetic regulation and detection, and immune and binding assays. In this work, we systematically analyzed the sequence and structure information of 199 bioluminescent and nonbioluminescent proteins, respectively. Based on the results, we presented a novel method called auto covariance of averaged chemical shift (acACS) for extracting structure features from a sequence. A classifier of support vector machine (SVM) fusing increment of diversity (ID) was used to distinguish bioluminescent proteins from nonbioluminescent proteins by combining dipeptide composition, reduced amino acid composition, evolutionary information, and acACS. The overall prediction accuracy evaluated by jackknife validation reached 82.16%. This result was better than that obtained by other existing methods. Improvement of the overall prediction accuracy reached up to 5.33% higher than those of the SVM and auto covariance of sequential evolution information by 10-fold cross-validation. The acACS algorithm also outperformed other feature extraction methods, indicating that our approach is better than other existing methods in the literature.

Ear Structure Feature Extraction Based on Multi-scale Hessian Matrix

In this paper, a new ear anatomy feature edge extraction method based on Hessian matrix is proposed. Stable edge is obtained from principal curvature image across scale space. Firstly, the side face image that includes an ear is filtered and forms Gaussian pyramid. Secondly, the 2D gray image in the pyramid was regarded as a surface, maximum and minimum principal curvature and their direction were calculated by using Hessian matrix, and principal curvature image was formed. The characteristic of surface is that gray level changes in edge area is sharp and the curvature is larger compared to that of the smooth area. In accordance with this characteristic, automatic hysteresis thresholding based on curvature direction flow is used to segment curvature images. Lastly, combine different scale threshold images to get the feature edge image. The experiments demonstrate that extracted feature edge is smooth and connected. New method is robust to noise, and is sensitive to the weak edge, using Hausdorff distance as similarity measurement of two edge images can obtain above 96% recognition rate.

Analysis and Extraction of Structural Features of Off-Line Handwritten Characters Based on Principal Curves

Analysis and Extraction of Structural Features of Off-Line Handwritten Characters Based on Principal Curves

WRITER IDENTIFICATION OF ARABIC TEXT USING STATISTICAL AND STRUCTURAL FEATURES

This article addresses writer identification of handwritten Arabic text. Several types of structural and statistical features were extracted from Arabic handwriting text. A novel approach was used to extract structural features that build on some of the main characteristics of the Arabic language. Connected component features for Arabic handwritten text as well as gradient distribution features, windowed gradient distribution features, contour chain code distribution features, and windowed contour chain code distribution features were extracted. A nearest neighbor (NN) classifier was used with the Euclidean distance measure. Data reduction algorithms (viz. principal component analysis [PCA], linear discriminant analysis [LDA], multiple discriminant analysis [MDA], multidimensional scaling [MDS], and forward/backward feature selection algorithm) were used. A database of 500 paragraphs handwritten in Arabic by 250 writers was used. The paragraphs used were randomly generated from a large corpus. NN provided the best accuracy in text-independent writer identification with top-1 result of 88.0%, top-5 result of 96.0%, and top-10 result of 98.5% for the first 100 writers. Extending the work to include all 250 writers and with the backward feature selection algorithm (using 54 out of 83 features), the system attained a top-1 result of 75.0%, top-5 result of 91.8%, and top-10 result of 95.4%.

Development of a brain MRI-based hidden Markov model for dementia recognition

BackgroundDementia is an age-related cognitive decline which is indicated by an early degeneration of cortical and sub-cortical structures. Characterizing those morphological changes can help to understand the disease development and contribute to disease early prediction and prevention. But modeling that can best capture brain structural variability and can be valid in both disease classification and interpretation is extremely challenging. The current study aimed to establish a computational approach for modeling the magnetic resonance imaging (MRI)-based structural complexity of the brain using the framework of hidden Markov models (HMMs) for dementia recognition.MethodsRegularity dimension and semi-variogram were used to extract structural features of the brains, and vector quantization method was applied to convert extracted feature vectors to prototype vectors. The output VQ indices were then utilized to estimate parameters for HMMs. To validate its accuracy and robustness, experiments were carried out on individuals who were characterized as non-demented and mild Alzheimer's diseased. Four HMMs were constructed based on the cohort of non-demented young, middle-aged, elder and demented elder subjects separately. Classification was carried out using a data set including both non-demented and demented individuals with a wide age range.ResultsThe proposed HMMs have succeeded in recognition of individual who has mild Alzheimer's disease and achieved a better classification accuracy compared to other related works using different classifiers. Results have shown the ability of the proposed modeling for recognition of early dementia.ConclusionThe findings from this research will allow individual classification to support the early diagnosis and prediction of dementia. By using the brain MRI-based HMMs developed in our proposed research, it will be more efficient, robust and can be easily used by clinicians as a computer-aid tool for validating imaging bio-markers for early prediction of dementia.

Employing Structural and Textual Feature Extraction for Semistructured Document Classification

This paper addresses XML document classification by considering both structural and content-based features of the documents. This approach leads to better constructing a set of informative feature vectors that represents both structural and textual aspects of XML documents. For this purpose, we integrate soft clustering of words and feature reduction into the process. To extract structural information, we employ an existing frequent tree-mining algorithm combined with an information gain filter to retrieve the most informative substructures from XML documents. However, for extracting content information, we propose soft clustering of words using each cluster as a textual feature. We have conducted extensive experiments on a benchmark dataset, namely 20NewsGroups, and an XML documents dataset given in LOGML that describes the web-server logs of user sessions. With regards to the classifier built only using our textual features, the results show that it outperforms a naive support-vector-machine (SVM)-based classifier, as well as an information retrieval classifier (IRC). We further demonstrate the effectiveness of incorporating both structural and content information into the process of learning, by comparing our classifier model and several XML document classifiers. In particular, by applying SVM and decision tree algorithms using our feature vector representation of XML documents dataset, we have achieved 85.79% and 87.04% classification accuracy, respectively, which are higher than accuracy achieved by XRules, a well-known structural-based XML document classifier.

Automating the Visualization of Biological Machines

With continued advances in three dimensional Electron Microscopy one is progressively able to elucidate and visualize natureˈs biological molecular machines (e.g. ribosomes, viruses) at varying resolutions and different conformations. I shall first discuss a suite of geometric optimization methods to automatically reconstruct a 3D volumetric model from a set of 2D images of the same object but with unknown pose. The solution to the inverse problem utilizes a pose determination algorithm, variational reconstruction of 3D scalar functions from 2D images, and geometry/image processing techniques of structure interpretation, feature extraction, and model refinement at progressively finer resolutions. Next, Ill briefly discuss molecular modeling techniques that enable dynamics, and visualization of structure‐function relationships of these biological machines.

Identification of Binding Specificity-Determining Features in Protein Families

We present a new approach for identifying features of ligand-protein binding interfaces that predict binding selectivity and demonstrate its effectiveness for predicting kinase inhibitor specificity. We analyzed a large set of human kinases and kinase inhibitors using clustering of experimentally determined inhibition constants (to define specificity classes of kinases and inhibitors) and virtual ligand docking (to extract structural and chemical features of the ligand-protein binding interfaces). We then used statistical methods to identify features characteristic of each class. Machine learning was employed to determine which combinations of characteristic features were predictive of class membership and to predict binding specificities and affinities of new compounds. Experiments showed predictions were 70% accurate. These results show that our method can automatically pinpoint on the three-dimensional binding interfaces pharmacophore-like features that act as "selectivity filters". The method is not restricted to kinases, requires no prior hypotheses about specific interactions, and can be applied to any protein families for which sets of structures and ligand binding data are available.

Noise tolerant local binary pattern operator for efficient texture analysis

The local binary pattern (LBP) operator is a very effective multi-resolution texture descriptor that can be applied in many image processing applications. However, existing LBP operators can not use the information of non-uniform patterns efficiently and they are also sensitive to noise. This paper proposes a noise tolerant extension of LBP operators to extract statistical and structural image features for efficient texture analysis. The proposed LBP operator uses a circular majority voting filter and suitable rotation-invariant labeling scheme to obtain more regular uniform and non-uniform patterns that have better discrimination ability and more robustness against noise. Experimental results on the Brodatz, CUReT and MeasTex databases show the improvement of the proposed LBP operator performance, especially when a large number of neighbors are used for extracting texture patterns.

Gait Recognition With Shifted Energy Image and Structural Feature Extraction

In this paper, we present a novel and efficient gait recognition system. The proposed system uses two novel gait representations, i.e., the shifted energy image and the gait structural profile, which have increased robustness to some classes of structural variations. Furthermore, we introduce a novel method for the simulation of walking conditions and the generation of artificial subjects that are used for the application of linear discriminant analysis. In the decision stage, the two representations are fused. Thorough experimental evaluation, conducted using one traditional and two new databases, demonstrates the advantages of the proposed system in comparison with current state-of-the-art systems.

Interactive Multiscale Tensor Reconstruction for Multiresolution Volume Visualization

Large scale and structurally complex volume datasets from high-resolution 3D imaging devices or computational simulations pose a number of technical challenges for interactive visual analysis. In this paper, we present the first integration of a multiscale volume representation based on tensor approximation within a GPU-accelerated out-of-core multiresolution rendering framework. Specific contributions include (a) a hierarchical brick-tensor decomposition approach for pre-processing large volume data, (b) a GPU accelerated tensor reconstruction implementation exploiting CUDA capabilities, and (c) an effective tensor-specific quantization strategy for reducing data transfer bandwidth and out-of-core memory footprint. Our multiscale representation allows for the extraction, analysis and display of structural features at variable spatial scales, while adaptive level-of-detail rendering methods make it possible to interactively explore large datasets within a constrained memory footprint. The quality and performance of our prototype system is evaluated on large structurally complex datasets, including gigabyte-sized micro-tomographic volumes.

Scene recognition combining structural and textural features

Automatic recognition of the contents of a scene is an important issue in the field of computer vision. Although considerable progress has been made, the complexity of scenes remains an important challenge to computer vision research. Most previous approaches for scene recognition are based on the so-called “bag of visual words” model, which uses clustering methods to quantize numerous local region descriptors into a codebook. The size of the codebook and the selection of initial clustering centers greatly affect the performance. Furthermore, the large size of the codebook leads to high computational costs and large memory consumption. To overcome these weaknesses, we present an unsupervised natural scene recognition approach that is not based on the “bag of visual words” model. This approach constructs multiple images of different resolutions and extracts structural and textural features from these images. The structural features are represented by weighted histograms of the gradient orientation descriptor, which is presented in this paper, and the textural features are represented by filter responses of Gabor filters and a Schmid set. We regard the structural and textural features as two independent feature channels, and combine them to realize automatic categorization of scenes using a support vector machine. We then evaluated our approach using three commonly used datasets with various scene categories. Our experiments demonstrate that the weighted histograms of the gradient orientation descriptor outperform the classical scale invariant feature transform descriptor in natural-scene recognition, and our approach achieves good performance with respect to current state-of-the-art methods.

Coupling of DEM and remote-sensing-based approaches for semi-automated detection of regional geostructural features in Zagros mountain, Iran

In recent years, remote-sensing data have increasingly been used for the interpretation of objects and mapping in various applications of engineering geology. Digital elevation model (DEM) is very useful for detection, delineation, and interpretation of geological and structural features. The use of image elements for interpretation is a common method to extract structural features. In this paper, linear features were extracted from the Landsat ETM satellite image and then DEM was used to enhance those objects using digital-image-processing filtering techniques. The extraction procedures of the linear objects are performed in a semi-automated way. Photographic elements and geotechnical elements are used as main keys to extract the information from the satellite image data. This paper emphasizes on the application of DEM and usage of various filtering techniques with different convolution kernel size applied on the DEM. Additionally, this paper discusses about the usefulness of DEM and satellite digital data for extraction of structural features in SW of Zagros mountain, Iran.

An Approach for Structural Feature Extraction for Distorted Tamil Character Recognition

Feature extraction is an important task for designing an OCR for recognizing degraded documents. Selection of a feature extraction method is probably the single most important factor in achieving high recognition performance in character recognition systems. Shape inconsistency among characters of the same structure is sometimes quite large because of the poor resources and environmental impact on the document images. Therefore, it is necessary to select features which can adapt to the shape variations irrespective of the distortions. Hence, in this paper, selection of appropriate standard structural features is taken as the primary task for various distortion types that are considered to recognize the Tamil distorted characters

An application using Wavelet coefficients for Video-content analysis and retrieval

The scheme of video-content analysis and indexing which involving structure analysis, abstraction, feature extraction and indexing poses many challenging research issues. This provides a unique opportunity for using intelligent media analysis techniques for creating advanced searching and browsing techniques to find relevant information quickly and inexpensively. This research mainly is to survey available and potential technologies for video monitoring and mining, the general methods of fast and efficient content-based analysis of video streams and to identify promising directions for research in this challenging area. This involves automatic detection of boundaries between the shots in a video and then those are indexed to form a library, saving the proper features of each shot/frame. This helps in the easy retrieval based on the shot according to the user requirements. We present an automation technique for video indexing and creation of a digital library, which is composed of stream shots and the wavelet coefficients for these shots. The wavelet coefficients are computed on the image and all the video frames for a full search function in all the frames of the indexed video and can be used for any number of shots/frames.

Offline handwritten Amharic word recognition

This paper describes two approaches for Amharic word recognition in unconstrained handwritten text using HMMs. The first approach builds word models from concatenated features of constituent characters and in the second method HMMs of constituent characters are concatenated to form word model. In both cases, the features used for training and recognition are a set of primitive strokes and their spatial relationships. The recognition system does not require segmentation of characters but requires text line detection and extraction of structural features, which is done by making use of direction field tensor. The performance of the recognition system is tested by a dataset of unconstrained handwritten documents collected from various sources, and promising results are obtained.

Unsupervised defect segmentation of patterned materials under NIR illumination

An unsupervised detection method for automatic flaw segmentation in patterned materials (textile, non-woven, paper) that has no need of any defect-free references or a training stage is presented in this paper. Printed materials having a pattern of colored squares, bands, etc. superimposed to the background texture can be advantageously analyzed using NIR illumination and a camera with enough sensitivity to this region of the spectrum. The contrast reduction of the pattern in the NIR image facilitates material inspection and defect segmentation. Underdetection and misdetection errors can be reduced in comparison with the inspection performed under visible illumination. For woven fabrics, with periodic structure, the algorithm is based on the structural feature extraction of the weave repeat from the Fourier transform of the sample image. These features are used to define a set of multiresolution bandpass filters adapted to the fabric structure that operate in the Fourier domain. Inverse Fourier transformation, binarization and merging of the information obtained at different scales lead to the output image that contains flaws segmented from the fabric background. For non-woven and random textured materials, the algorithm combines the multiresolution Gabor analysis of the sample image with a statistical analysis of the wavelet coefficients corresponding to each detail. The information of all the channels is merged in a single binary output image where the defect appears segmented from the background. The method is applicable to random, non-periodic, and periodic textures. Since all the information to inspect a sample is obtained from the sample itself, the method is proof against heterogeneities between different samples of the material, in-plane positioning errors, scale variations and lack of homogeneous illumination. Experimental results are presented for a variety of materials and defects.