Feature Extraction Fusion Research Articles

Design of Low-Cost Personal Identification System That Uses Combined Palm Vein and Palmprint Biometric Features

This paper proposes a low-cost personal identification system that uses the combined palm vein and palmprint biometric features. The system consists of near-infrared and visible light-emitting diode (LED) arrays, a low-cost visual sensor, a Xilinx chip, and other components. A real-time image quality assessment (IQA) method for the combined palm vein and palmprint biometric features is also proposed. Two types of the LED with central frequency spectra of 890 and 680 nm are used to capture the palm vein and palmprint, respectively. The adaptive feedback control of the diode brightness is in accordance with the image quality assessed by the combined 2D entropy and local 2D entropy. The palm vein and palmprint images are acquired nearly simultaneously, and each acquired image undergoes a few preprocessing steps for extraction of the vein and print patterns. We use an image-level wavelet-based fusion strategy to reduce image storage requirement for the embedded platform and implement a complex wavelet-based fusion strategy for the PC platform. A deep scattering convolutional network is applied for extracting the features of the fused images, and a multi-class support vector machine is used for training and recognition. Characteristics of some vision-based personal identification systems are discussed. The proposed real-time IQA method with fusion strategy and feature extraction algorithm in our prototype system has substantially less operational requirements than that of the previous fusion strategies. It also demands less memory and yields lower equal error rate than the classical feature extraction algorithms.

Fusion Feature Extraction Based on Auditory and Energy for Noise-Robust Speech Recognition

Environmental noise can pose a threat to the stable operation of current speech recognition systems. It is therefore essential to develop a front feature set that is able to identify speech under low signal-to-noise ratio. In this paper, a robust fusion feature is proposed that can fully characterize speech information. To obtain the cochlear filter cepstral coefficients (CFCC), a novel feature is first extracted by the power-law nonlinear function, which can simulate the auditory characteristics of the human ear. Speech enhancement technology is then introduced into the front end of feature extraction, and the extracted feature and their first-order difference are combined in new mixed features. An energy feature Teager energy operator cepstral coefficient (TEOCC) is also extracted, and combined with the above-mentioned mixed features to form the fusion feature sets. Principal component analysis (PCA) is then applied to feature selection and optimization of the feature set, and the final feature set is used in a non-specific persons, isolated words, and small-vocabulary speech recognition system. Finally, a comparative experiment of speech recognition is designed to verify the advantages of the proposed feature set using a support vector machine (SVM). The experimental results show that the proposed feature set not only display a high recognition rate and excellent anti-noise performance in speech recognition, but can also fully characterize the auditory and energy information in the speech signals.

Affine invariant fusion feature extraction based on geometry descriptor and BIT for object recognition

It is difficult to recognise an image with affine transformation due to viewing angle and distance variations. Therefore, affine invariant feature extraction is a valuable technology in the field of image recognition. Inspired by bio-visual mechanism, an affine invariant for object recognition method based on a fusion feature framework is proposed in this study, which employs geometry descriptor and double biologically inspired transformation (DBIT). First, a shape feature of interest detector is adopted to detect contour features. Then, the area estimation of affine region detector is utilised to construct area ratio feature vectors. Second, an orientation edge detector is built to highlight the edges of different directions. On this basis, local space frequency detector is adopted to measure the spatial frequency at each direction and interval, which converts the output map into DBIT feature vectors. A weighted fusion strategy is performed based on Pearson correlation distance to fuse the geometry feature and DBIT feature. Some tests for Alphanumeric, Coil-100 MPEG-7, Mixed National Institute of Standards and Technology (MNIST) and Olivetti Research Laboratory face images database (ORL) database remain highly stable recognition accuracy, even when the shear factor is between -0.5 and + 0.5. The experiment results show the authors' proposed approach has a nice performance in feature invariance, selectivity and recognition accuracy.

A Fusion Feature Extraction Method Using EEMD and Correlation Coefficient Analysis for Bearing Fault Diagnosis

Acceleration sensors are frequently applied to collect vibration signals for bearing fault diagnosis. To fully use these vibration signals of multi-sensors, this paper proposes a new approach to fuse multi-sensor information for bearing fault diagnosis by using ensemble empirical mode decomposition (EEMD), correlation coefficient analysis, and support vector machine (SVM). First, EEMD is applied to decompose the vibration signal into a set of intrinsic mode functions (IMFs), and a correlation coefficient ratio factor (CCRF) is defined to select sensitive IMFs to reconstruct new vibration signals for further feature fusion analysis. Second, an original feature space is constructed from the reconstructed signal. Afterwards, weights are assigned by correlation coefficients among the vibration signals of the considered multi-sensors, and the so-called fused features are extracted by the obtained weights and original feature space. Finally, a trained SVM is employed as the classifier for bearing fault diagnosis. The diagnosis results of the original vibration signals, the first IMF, the proposed reconstruction signal, and the proposed method are 73.33%, 74.17%, 95.83% and 100%, respectively. Therefore, the experiments show that the proposed method has the highest diagnostic accuracy, and it can be regarded as a new way to improve diagnosis results for bearings.

Deriving High Spatiotemporal Remote Sensing Images Using Deep Convolutional Network

Due to technical and budget limitations, there are inevitably some trade-offs in the design of remote sensing instruments, making it difficult to acquire high spatiotemporal resolution remote sensing images simultaneously. To address this problem, this paper proposes a new data fusion model named the deep convolutional spatiotemporal fusion network (DCSTFN), which makes full use of a convolutional neural network (CNN) to derive high spatiotemporal resolution images from remotely sensed images with high temporal but low spatial resolution (HTLS) and low temporal but high spatial resolution (LTHS). The DCSTFN model is composed of three major parts: the expansion of the HTLS images, the extraction of high frequency components from LTHS images, and the fusion of extracted features. The inputs of the proposed network include a pair of HTLS and LTHS reference images from a single day and another HTLS image on the prediction date. Convolution is used to extract key features from inputs, and deconvolution is employed to expand the size of HTLS images. The features extracted from HTLS and LTHS images are then fused with the aid of an equation that accounts for temporal ground coverage changes. The output image on the prediction day has the spatial resolution of LTHS and temporal resolution of HTLS. Overall, the DCSTFN model establishes a complex but direct non-linear mapping between the inputs and the output. Experiments with MODerate Resolution Imaging Spectroradiometer (MODIS) and Landsat Operational Land Imager (OLI) images show that the proposed CNN-based approach not only achieves state-of-the-art accuracy, but is also more robust than conventional spatiotemporal fusion algorithms. In addition, DCSTFN is a faster and less time-consuming method to perform the data fusion with the trained network, and can potentially be applied to the bulk processing of archived data.

Hashing Based Hierarchical Feature Representation for Hyperspectral Imagery Classification

Integrating spectral and spatial information is proved effective in improving the accuracy of hyperspectral imagery classification. In recent studies, two kinds of approaches are widely investigated: (1) developing a multiple feature fusion (MFF) strategy; and (2) designing a powerful spectral-spatial feature extraction (FE) algorithm. In this paper, we combine the advantages of MFF and FE, and propose an ensemble based feature representation method for hyperspectral imagery classification, which aims at generating a hierarchical feature representation for the original hyperspectral data. The proposed method is composed of three cascaded layers: firstly, multiple features, including local, global and spectral, are extracted from the hyperspectral data. Next, a new hashing based feature representation method is proposed and conducted on the features obtained in the first layer. Finally, a simple but efficient extreme learning machine classifier is employed to get the classification results. To some extent, the proposed method is a combination of MFF and FE: instead of feature fusion or single feature extraction, we use an ensemble strategy to provide a hierarchical feature representation for the hyperspectral data. In the experiments, we select two popular and one challenging hyperspectral data sets for evaluation, and six recently proposed methods are compared. The proposed method achieves respectively 89.55%, 99.36% and 77.90% overall accuracies in the three data sets with 20 training samples per class. The results prove that the performance of the proposed method is superior to some MFF and FE based ones.

An Image based Diagnostic System for Lung Disease Classification

Model-based detection and classification of nodules are two major steps in CAD systems design and evaluation. A common health problem, lung diseases are the most prevailing medical conditions throughout the world. In this paper, Lung diseases are automatically classified as Emphysema, Bronchitis, Pleural effusion and normal lung.The lung CT images are taken as input, preprocessing is applied, feature extraction is done by various methods such as Gabor filter extracts the texture features, walsh hadamard transform extracts the pixel co-efficient values, and a fusion method is proposed in this work which extracts the median absolute deviation values. Feature selection including statistical correlation based methods and Genetic Algorithm for searching in feature vector space are investigated. Four types of the classifiers are used where the Multi-Layer Perceptron Neural Network (MLP-NN) classifier with proposed fusion feature extraction method, genetic algorithm feature selection method gives promising result of 91% accuracy than J48, K- Nearest Neighbour and Naïve bayes classifiers.

A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification

The sensitive feature extraction from vibration signals is still a great challenge for effective fault classification of rolling element bearing. Current fault classification generally depends on feature pattern difference of different fault classes. This paper explores the active role of healthy pattern in fault classification and proposes a new fusion feature extraction method based on locality preserving projections (LPP). This study intends to discover the local feature pattern difference between each bearing status and the healthy condition to characterize and discriminate different bearing statuses. Specifically, the proposed fusion feature is achieved by two main steps. In the first step, a two-class model is firstly constructed for each class by using this class of signals and healthy condition signals. Then a fusion mapping is generated by mathematically combing the mappings of the LPP or its improvement for all two-class models. In the second step, the LPP is further applied to reduce the fusion mapping dimension, which is to find more sensitive low-dimensional information hidden in the high-dimensional fusion feature structure. The final achieved fusion feature can enhance the discrimination between all classes by improving the between-class scatter and within-class scatter for fault classification. Experimental results using different bearing fault types and severities under different loads show that the proposed method is well-suited and effective for bearing fault classification.

An Advanced Vision System for the Automatic Inspection of Corrosions on Pole Tips in Hard Disk Drives

Pole tips of hard disk drives are made of FeCo film, which is prone to corrosion due to the corrosive nature of the environment and wide range of pH experienced during the production. A machine vision system is utilized for the automatic auditing of corroded pole tips. This comprises three steps: 1) extraction of the top shield region; 2) fusion of extracted features; and 3) decision making. This model is capable of detecting corrosion in the critical part of the pole tips. The experiments show that the proposed system detects corrosion with high accuracy, whereas the overall processing time meets industrial requirements.

Cancelable Fusion of Face and Ear for Secure Multi-Biometric Template

Biometric fusion to achieve multimodality has emerged as a highly successful new approach to combat problems of unimodal biometric system such as intraclass variability, interclass similarity, data quality, non-universality, and sensitivity to noise. The authors have proposed new type of biometric fusion called cancelable fusion. The idea behind the cancelable biometric or cancelability is to transform a biometric data or feature into a new one so that the stored biometric template can be easily changed in a biometric security system. Cancelable fusion does the fusion of multiple biometric trait in addition it preserve the properties of cancelability. In this paper, the authors present a novel architecture for template generation within the context of the cancelable multibiometric fusion. The authors develop a novel cancelable biometric template generation algorithm using cancelable fusion, random projection and transformation-based feature extraction and selection. The authors further validate the performance of the proposed algorithm on a virtual multimodal face and ear database.

How can data fusion help humanitarian mine action?

In this article, we present our views regarding multi-sensor data fusion potentials to help two types of humanitarian mine action: close-in detection and mined area reduction. Several approaches are discussed, reflecting our thoughts and experience in this extremely sensitive field of application. For close-in detection, our work on modelling and fusion of extracted features is based on two main methods, one related to the belief function theory and the other one to the possibility theory. The approaches are tested using real data coming from three complementary sensors (ground-penetrating radar, infrared sensor, and metal detector), collected within the Dutch HOM-2000 project. These results are obtained within two Belgian humanitarian demining projects, HUmanitarian DEMining (HUDEM) and Belgian Mine Action Technology (BEMAT). In the case of mined area reduction, our multi-sensor data fusion methods are based on the belief function theory and on the fuzzy logic and applied to real data of synthetic-aperture radar and multi-spectral sensors, gathered within the EU project on Space and Airborne Mined Area Reduction Tools (SMART). The importance of including various knowledge sources is discussed too.