Feature-based Techniques Research Articles

Blur and occlusion invariant palmprint recognition with block-wise local phase quantization histogram

A method for palmprint verification and identification, which is robust to blurred and occluded palmprints is proposed. An algorithm is proposed for region of interest (ROI) extraction that aligns each palmprint ROI on the same position. First-level decomposition of the ROI using a Haar wavelet gives an approximation ROI (AROI) and reduces computational overhead. Local phase quantization (LPQ) based on quantization of phase information by local Fourier transform is applied on AROI to obtain blur invariant palmprint features. The LPQ image is divided into M×M nonoverlapping blocks, and histograms of these blocks are considered as features. The entropy of each block is used to identify it as occluded or nonoccluded. The average chi-square distance between corresponding nonoccluded blocks is calculated for measuring similarity. Experiments performed on PolyU 2D, CASIA, IITD, and IIITDMJ palmprint databases show that the proposed system is independent of acquisition device. Moreover, to justify the robustness of the proposed system, experiments are performed with variable amounts of blur and occlusion present in test palmprints. The performance of the proposed system is compared with several local feature-based palmprint recognition techniques on normal as well as occluded palmprints.

Spectral Feature Detection With Sub-Nyquist Sampling for Wideband Spectrum Sensing

Compressive sensing (CS) has been successfully applied to alleviate the sampling bottleneck in wideband spectrum sensing leveraging the sparsity described by the low spectral occupancy of the licensed radios. However, the existence of interferences emanating from low-regulated transmissions, which cannot be taken into account in the CS model because of their non-regulated nature, greatly degrade the identification of licensed activity. This paper presents a feature-based technique for primary user's spectrum identification with interference immunity which works with a reduced amount of data. The proposed method not only detects which frequencies are occupied by primary users' but also identifies the primary users' transmitted power. The basic strategy is to compare the a priori known spectral shape of the primary user with the power spectral density of the received signal. This comparison is made in terms of autocorrelation by means of a correlation matching, thus avoiding the computation of the power spectral density of the received signal. The essence of the novel interference rejection mechanism lies in preserving the positive semidefinite character of the residual correlation, which is inserted by means of a weighted formulation of the $l_1$ -minimization. Simulation results show the effectiveness of the technique for interference suppression and primary user detection.

Modeling citizens’ urban time-use using adaptive hypermedia surveys to obtain an urban planning, citizen-centric, methodological reinvention

Traditional urban planning methodologies cannot solve today’s planning challenges that cities pose. Citizens’ needs have substantially changed the last decade while urban planning methods have remained the same since the 70s. A more citizen-centric urban planning is presented in this paper. To that end, and to effectively deal with citizens’ larger data needed for the urban planning methodological reinvention, interdisciplinary work is supported with information and communication technologies. We conducted a survey of citizens on urban time-use (Marsal MLl and López MB (2014) Smart urban planning: Designing urban-land-use from urban time-use. Journal of Urban Technology 21(1): 39–54). The results on urban-time-use distribution were converted into urban-land-uses as pioneering methodology for the reinvention of urban planning methodologies. In order to get the highest voluntary participation from citizens to ensure a good representation of all ages and social groups, the survey has to be designed with adaptive hypermedia techniques. The adaptive hypermedia techniques we propose in this paper combine stereotype and feature-based models which we explore for the purpose to include them in the survey. The combination of stereotype and feature-based models has different advantages, among others: stereotype techniques avoid to initiate survey profiles from scratch and feature-based techniques allow a personalized questionnaire to be employed. Moreover, personalization, in combination with user profiles, allows prediction which is of great interest for this research due to its planning purposes.

Feature-based multiexposure image-sequence fusion with guided filter and image alignment

Multiexposure fusion images have a higher dynamic range and reveal more details than a single captured image of a real-world scene. A clear and intuitive feature-based fusion technique for multiexposure image sequences is conceptually proposed. The main idea of the proposed method is to combine three image features [phase congruency (PC), local contrast, and color saturation] to obtain weight maps of the images. Then, the weight maps are further refined using a guided filter which can improve their accuracy. The final fusion result is constructed using the weighted sum of the source image sequence. In addition, for multiexposure image-sequence fusion involving dynamic scenes containing moving objects, ghost artifacts can easily occur if fusion is directly performed. Therefore, an image-alignment method is first used to adjust the input images to correspond to a reference image, after which fusion is performed. Experimental results demonstrate that the proposed method has a superior performance compared to the existing methods.

Suitable models for face geometry normalization in facial expression recognition

Recently, facial expression recognition has attracted much attention in machine vision research because of its various applications. Accordingly, many facial expression recognition systems have been proposed. However, the majority of existing systems suffer from a critical problem: geometric variability. It directly affects the performance of geometric feature-based facial expression recognition approaches. Furthermore, it is a crucial challenge in appearance feature-based techniques. This variability appears in both neutral faces and facial expressions. Appropriate face geometry normalization can improve the accuracy of each facial expression recognition system. Therefore, this paper proposes different geometric models or shapes for normalization. Face geometry normalization removes geometric variability of facial images and consequently, appearance feature extraction methods can be accurately utilized to represent facial images. Thus, some expression-based geometric models are proposed for facial image normalization. Next, local binary patterns and local phase quantization are used for appearance feature extraction. A combination of an effective geometric normalization with accurate appearance representations results in more than a 4% accuracy improvement compared to several state-of-the-arts in facial expression recognition. Moreover, utilizing the model of facial expressions which have larger mouth and eye region sizes gives higher accuracy due to the importance of these regions in facial expression.

Image Stitching System Based on ORB Feature-Based Technique and Compensation Blending

The construction of a high-resolution panoramic image from a sequence of input overlapping images of the same scene is called image stitching/mosaicing. It is considered as an important, challenging topic in computer vision, multimedia, and computer graphics. The quality of the mosaic image and the time cost are the two primary parameters for measuring the stitching performance. Therefore, the main objective of this paper is to introduce a high-quality image stitching system with least computation time. First, we compare many different features detectors. We test Harris corner detector, SIFT, SURF, FAST, GoodFeaturesToTrack, MSER, and ORB techniques to measure the detection rate of the corrected keypoints and processing time. Second, we manipulate the implementation of different common categories of image blending methods to increase the quality of the stitching process. From experimental results, we conclude that ORB algorithm is the fastest, more accurate, and with higher performance. In addition, Exposure Compensation is the highest stitching quality blending method. Finally, we have generated an image stitching system based on ORB using Exposure Compensation blending method.

Parallax Effect Free Mosaicing of Underwater Video Sequence Based on Texture Features

In this paper, we present feature-based technique for construction of mosaic image from underwater video sequence, which suffers from parallax distortion due to propagation properties of light in the underwater environment. The most of the available mosaic tools and underwater image mosaicing techniques yields final result with some artifacts such as blurring, ghosting and seam due to presence of parallax in the input images. The removal of parallax from input images may not reduce its effects instead it must be corrected in successive steps of mosaicing. Thus, our approach minimizes the parallax effects by adopting an efficient local alignment technique after global registration. We extract texture features using Centre Symmetric Local Binary Pattern (CS-LBP) descriptor in order to find feature correspondences, which are used further for estimation of homography through RANSAC. In order to increase the accuracy of global registration, we perform preprocessing such as colour alignment between two selected frames based on colour distribution adjustment. Because of existence of 100% overlap in consecutive frames of underwater video, we select frames with minimum overlap based on mutual offset in order to reduce the computation cost during mosaicing. Our approach minimizes the parallax effects considerably in final mosaic constructed using our own underwater video sequences.

Enhanced Fast Robustness Digital Image Watermarking Scheme with Secure Chaotic Encryption Scheme Via Selecting Optimal Feature Points

In recent years, with the rapid development of internet, digital data has been widely used in various science and engineering applications. But, this development has also led to several threats and attacks from the intruders. Most of the organizations are very much concerned about the Illegal copyright activities which have become a series issue. A number of copyright protection schemes have been developed in the literature. Watermarking is one of the best known methods for copyright protection. Earlier Bit Error Rate (BER) is been observed to provide good results for choosing the optimal feature region from the watermarked image. However it fails to produce more accurate image quality measures. This research work mainly focuses on a novel technique implemented based on the simulated attacking approach and the GA-based Multi Dimensional Knapsack Problem (MDKP) solving technique to select the most sufficient feature regions for robust digital image watermarking without affecting the image quality. A major limitation in existing methods is the security problems which lead to decrease in the integrity of digital watermarking scheme. With that problem, the definite optimization problems (variant problems) cannot be resolved using genetic algorithms. To focus on the above mentioned problems, the proposed work uses two approaches. Chaotic Iterative Process encryption scheme is used to solve the security problems for watermarking image. Next, the optimal feature points can be chosen from the watermarked image by Modified version of Peak Signal Noise Ratio (MPSNR). MPSNR is based on the enhanced quality measurement of which point opposes the predefined and undefined attacks. With those two approaches the PSO algorithm is employed as a search strategy for identifying an optimal feature subset. The experimental results illustrates that comparing with various well-known feature-based techniques, the proposed method shows better performance in robust and more secure digital watermarking.

MRI upsampling using feature-based nonlocal means approach.

In magnetic resonance imaging (MRI), spatial resolution is limited by several factors such as acquisition time, short physiological phenomena, and organ motion. The acquired image usually has higher resolution in two dimensions (the acquisition plane) in comparison with the third dimension, resulting in highly anisotropic voxel size. Interpolation of these low resolution (LR) images using standard techniques, such as linear or spline interpolation, results in distorted edges in the planes perpendicular to the acquisition plane. This poses limitation on conducting quantitative analyses of LR images, particularly on their voxel-wise analysis and registration. We have proposed a new non-local means feature-based technique that uses structural information of a high resolution (HR) image with a different contrast and interpolates the LR image. In this approach, the similarity between voxels is estimated using a feature vector that characterizes the laminar pattern of the brain structures, resulting in a more accurate similarity measure in comparison with conventional patch-based approach. This technique can be applied to LR images with both anisotropic and isotropic voxel sizes. Experimental results conducted on brain MRI scans of patients with brain tumors, multiple sclerosis, epilepsy, as well as schizophrenic patients and normal controls show that the proposed method is more accurate, requires fewer computations, and thus is significantly faster than a previous state-of-the-art patch-based technique. We also show how the proposed method may be used to upsample regions of interest drawn on LR images.

Syntactic and Sentence Feature Based Hybrid Approach for Text Summarization

Recently, there has been a significant research in automatic text summarization using feature- based techniques in which most of them utilized any one of the soft computing techniques. But, making use of syntactic structure of the sentences for text summarization has not widely applied due to its difficulty of handling it in summarization process. On the other hand, feature-based technique available in the literature showed efficient results in most of the techniques. So, combining syntactic structure into the feature-based techniques is surely smooth the summarization process in a way that the efficiency can be achieved. With the intention of combining two different techniques, we have presented an approach of text summarization that combines feature and syntactic structure of the sentences. Here, two neural networks are trained based on the feature score and the syntactic structure of sentences. Finally, the two neural networks are combined with weighted average to find the sentence score of the sentences. The experimentation is carried out using DUC 2002 dataset for various compression ratios. The results showed that the proposed approach achieved F-measure of 80% for the compression ratio 50 % that proved the better results compared with the existing techniques.

Damage Characterization in Frame Structures using Output-only Modal and Feature-based Techniques

A combined approach involving a modal-based as well as feature-based technique has been proposed for damage characterization in frame structures. In modal-based approach, the modal parameters of a structure are estimated from ambient responses using an output- only system identification technique and subsequently the fundamental mode shape and its derivatives are used for damage characterization purpose. In feature-based approach, autoregressive models with exogenous input (ARX model) are built up using healthy and damaged structural responses and the first ARX model coefficients are used for damage characterization. Simulation studies are carried out on a two-dimensional three-bay twelve- story steel moment-resisting frame to illustrate the applicability of the techniques in identifying damages in beams and columns. Both modal and feature-based approaches are capable of identifying the nature and extent of damages. However, an improvement of the algorithm is desirable for damage quantification and prognosis.

Statistical comparison of classifiers for script identification from multi-script handwritten documents

Script identification for handwritten document image is an open document analysis problem especially for multilingual optical character recognition (OCR) system. To design the OCR system for multi-script document pages, it is essential to recognise different scripts before running a particular OCR system of a script. The present work reports an intelligent feature-based technique for word-level script identification in multi-script handwritten document pages. At first, the text lines and then the words are extracted from the document pages. A set of 39 distinctive features have been designed of which eight features are topological and the rest (31) are based on convex hull for each word image. For selection of a suitable classifier, performances of multiple classifiers are evaluated with the designed feature set on multiple subsets of freely available database CMATERdb1.5.1 (http://www.code.google.com/p/cmaterdb), which comprises of 150 handwritten document pages containing both Devnagari and Roman script words. Statistical significance tests on these performance measures declare MLP to be the best performing one. The overall word-level script identification accuracy with MLP classifier on the said database is observed as 99.74%.

Classification of Chinese Popular Songs Using a Fusion Scheme of GMM Model Estimate and Formant Feature Analysis

In this paper, a fusion scheme that combines Gaussian mixture model (GMM) calculations and formant feature analysis, called GMM-Formant, is proposed for classification of Chinese popular songs. Generally, automatic classification of popular music could be performed by two main categories of techniques, model-based and feature-based approaches. In model-based classification techniques, GMM is widely used for its simplicity. In feature-based music recognition, the formant parameter is an important acoustic feature for evaluation. The proposed GMM-Formant method takes use of linear interpolation for combining GMM likelihood estimates and formant evaluation results appropriately. GMM-Formant will effectively adjust the likelihood score, which is derived from GMM calculations, by referring to certain degree of formant feature evaluation outcomes. By considering both model-based and feature-based techniques for song classification, GMM-Formant provides a more reliable recognition classification result and therefore will maintain a satisfactory performance in recognition accuracy. Experimental results obtained from a musical data set of numerous Chinese popular songs show the superiority of the proposed GMM-Formant. Keywords: Song classification; Gaussian mixture model; Formant feature; GMM-Formant.

A new sketch-based 3D model retrieval approach by using global and local features

With the rapid growth of available 3D models, fast retrieval of suitable 3D models has become a crucial task for industrial applications. This paper proposes a novel sketch-based 3D model retrieval approach which utilizes both global feature-based and local feature-based techniques. Unlike current approaches which use either global or local features, as well as do not take into account semantic relations between local features, we extract these two kinds of feature information from the representative 2D views of 3D models that can facilitate semantic description and retrieval for 3D models. Global features represent the gross exterior boundary shape information, and local features describe the interior details by compact visual words. Specifically, an improved bag-of-features method is provided to extract local features and their latent semantic relations. In addition, an efficient two-stage matching strategy is used to measure the distance between the query sketch and 3D models for selection and refinement. Experiment results demonstrate that our approach which combines these two kinds of complementary features significantly outperforms several state-of-the-art approaches.

Improved process monitoring and supervision based on a reliable multi-stage feature-based pattern recognition technique

This paper investigates two new multisensor data fusion algorithms for object detection in monitoring of industrial processes. The goals were to reduce the rate of false detection and obtain reliable decisions on the presence of target objects. The monitoring system uses acceleration sensors and is used as a sensor-cluster. In principle the approach can include arbitrary data acquisition techniques. Two approaches were proposed. The first uses a short-time Fourier transform (STFT) as a prefilter to extract relevant features from the acceleration signals. The features extracted from different sensor channels are first classified using support vector machine (SVM)-based filters. A novel decision fusion process to combine individual decisions was developed. The second approach uses a continuous wavelet transform (CWT) as a prefilter to extract relevant features from the acceleration signals. The features extracted from different sensor signals are subjected to further prefiltering processes before SVM-based classification. The individual decision functions are then combined in a decision fusion module. The classification system was trained and validated using real industrial data. The two approaches were tested using the same data and their performance and modeling complexity are compared. The developed approaches show strong improvements in detection and false alarm rates.

Geodesics-Based Image Registration: Applications To Biological And Medical Images Depicting Concentric Ring Patterns

In many biological or medical applications, images that contain sequences of shapes are common. The existence of high inter-individual variability makes their interpretation complex. In this paper, we address the computer-assisted interpretation of such images and we investigate how we can remove or reduce these image variabilities. The proposed approach relies on the development of an efficient image registration technique. We first show the inadequacy of state-of-the-art intensity-based and feature-based registration techniques for the considered image datasets. Then, we propose a robust variational method which benefits from the geometrical information present in this type of images. In the proposed non-rigid geodesics-based registration, the successive shapes are represented by a level-set representation, which we rely on to carry out the registration. The successive level sets are regarded as elements in a shape space and the corresponding matching is that of the optimal geodesic path. The proposed registration scheme is tested on synthetic and real images. The comparison against results of state-of-the-art methods proves the relevance of the proposed method for this type of images.

Energy Management of Planetary Rovers Using a Fast Feature-Based Path Planning and Hardware-in-the-Loop Experiments

This paper presents a novel feature-based technique for path optimization problems, in which the performance index is defined to minimize the energy consumption of a rover with consideration of terrain, kinematic, and dynamic constraints. The proposed method estimates rover energy consumption by discretizing a path and by extracting statistical data for fast calculation of the performance index. The concepts of grouped data and data discretization techniques are used to analyze the energy-related data obtained from the search environment. The method improves runtime computation by statistically calculating the energy consumption of a rover for a defined path, rather than solving the dynamic equations of the rover. This technique is computationally more efficient than other energy optimization approaches when it estimates rover energy consumption with sufficient accuracy. The Genetic Algorithm (GA) solver is integrated to the approach to illustrate the efficiency of the algorithm. Additionally, a hardware-in-the-loop (HIL) simulation is developed for the validation of the rover's power flow as it traverses through the optimal path by incorporating rover hardware components within real-time simulation.

Feature-based classification of amino acid substitutions outside conserved functional protein domains.

There are more than 500 amino acid substitutions in each human genome, and bioinformatics tools irreplaceably contribute to determination of their functional effects. We have developed feature-based algorithm for the detection of mutations outside conserved functional domains (CFDs) and compared its classification efficacy with the most commonly used phylogeny-based tools, PolyPhen-2 and SIFT. The new algorithm is based on the informational spectrum method (ISM), a feature-based technique, and statistical analysis. Our dataset contained neutral polymorphisms and mutations associated with myeloid malignancies from epigenetic regulators ASXL1, DNMT3A, EZH2, and TET2. PolyPhen-2 and SIFT had significantly lower accuracies in predicting the effects of amino acid substitutions outside CFDs than expected, with especially low sensitivity. On the other hand, only ISM algorithm showed statistically significant classification of these sequences. It outperformed PolyPhen-2 and SIFT by 15% and 13%, respectively. These results suggest that feature-based methods, like ISM, are more suitable for the classification of amino acid substitutions outside CFDs than phylogeny-based tools.

Reply to “Comments on ‘Principal Component Analysis of Single-Beam Echo-Sounder Signal Features for Seafloor Classification’” by J. M. Preston and B. R. Biffard

In the paper published in the IEEE Journal of Oceanic Engineering, vol. 36, no. 2, pp. 259-271, Apr. 2011, results of applying a feature-based classification technique to single-beam echo-sounder data were presented. Comments on this contribution were presented in a correspondence paper published in the IEEE, Journal of Oceanic Engineering, vol. 37, no. 4, pp. 764-765, Oct. 2012). In this reply, the items addressed in the comments are reflected upon.

A survey of control-chart pattern-recognition literature (1991–2010) based on a new conceptual classification scheme

Control Chart Pattern Recognition (CCPR) is a critical task in Statistical Process Control (SPC). Abnormal patterns exhibited in control charts can be associated with certain assignable causes adversely affecting the process stability. Abundant literature treats the detection of different Control Chart Patterns (CCPs). In fact, numerous CCPR studies have been developed according to various objectives and hypotheses. Despite the widespread literature on this topic, efforts to review and analyze research on CCPR are very limited. For this reason, this survey paper proposes a new conceptual classification scheme, based on content analysis method, to classify past and current developments in CCPR research. More than 120 papers published on CCPR studies within 1991–2010 were classified and analyzed. Major findings of this survey include the following. (1) The most popular CCPR studies deal with independently and identically distributed process data. (2) Some recent studies on identification of mean shifts or/and variance shifts of a multivariate process are based on innovative techniques. (3) The percentage of studies that address concurrent pattern identification is increasing. (4) The majority of the reviewed articles use Artificial Neural Network (ANN) approach. Feature-based techniques, in particular wavelet-denoise, are investigated for improving the recognition performance of ANN. For the same reason, there is a general trend followed by many authors who propose hybrid, modular and integrated ANN recognizer designs combined with decision tree learning, particle swarm optimization, etc. (5) There are two main categories of performance criteria used to evaluate CCPR approaches: statistical criteria that are related to two conventional Average Run Length (ARL) measures, and recognition-accuracy criteria, which are not based on these ARL measures. The most applied criteria are recognition-accuracy criteria, mainly for ANN-based approaches. Performance criteria which are related to ARL measures are insufficient and inappropriate in the case of concurrent pattern identification. Finally, this paper briefly discusses some future research directions and our perspectives.