Test Feature Vectors Research Articles

An Efficient Method for Similarity Measure in Independent PCA based Classification

The paper presents a new approach of finding nearest neighbor in image classification algorithm by proposing efficient method for similarity measure. Generally in supervised classification, after finding the feature vectors of training images and testing images, nearest neighbor classifier does the classification job. This classifier uses different distance measures such as Euclidean distance, Manhattan distance etc. to find the nearest training feature vector. This paper proposes to use Mean Squared Error (MSE) to find the nearness between two images. Initially Independent Principal Component Analysis (PCA),which we discussed in our earlier work, is applied to images of each class to generate Eigen coordinate system for that class. Then for the given test image, a set of feature vectors is generated. New images are reconstructed using each Eigen coordinate system and the corresponding test feature vector. Lowest MSE between the given test image and new reconstructed image indicates the corresponding class for that image. The experiments are conducted on COIL-100 database. The performance is also compared with distance based nearest neighbor classifier. Results show that the proposed method achieves high accuracy even for small size of training set.

Sparse Representation-Based Heartbeat Classification Using Independent Component Analysis

The classification of heartbeats is crucial to identify an arrhythmia. This paper proposes a new method that combines independent component analysis (ICA) with sparse representation-based classification (SRC) to distinguish eight types of heartbeats. We use ICA to extract useful features from heartbeats. A feature vector consists of 100 ICA features along with a RR interval. We use SRC to compute a sparse representation of a test feature vector with respect to all training feature vectors. The type of a test feature vector is determined using the concentration degree of sparse coefficients on each heartbeat type. For experimental purposes, 9800 heartbeats are extracted from the MIT-BIH electrocardiogram (ECG) database. The results show that our proposed method performs better than conventional methods, with 98.35% accuracy and 94.49%-100% sensitivities to several heartbeat types.

Face recognition based on ridgelet transforms

This paper relates to a face recognition and verification technique based on ridgelet transforms. Our proposed method first uses the ridgelet transform of the face image for feature extraction. This involves first applying a sequen- tial combination of radon and wavelet transforms to both the training and test images. The result is then decomposed into a set of feature vectors. The Euclidean distance between training and test feature vectors is finally used for the actual recognition. Before applying any transform on the training image, we normalize the image using a segmentation process based on the YCbCr colour model. This essentially detects the largest region of skin in the image. Experimental results using Yale, AT&T, faces94, faces96 and Indian databases show the superiority of the proposed method with some of the existing popular algorithms.

Machine fault diagnosis based on Gaussian mixture model and its application

This paper presents a simple and efficient machine fault diagnosis approach based on Gaussian mixture model (GMM). After feature vectors that represent different machine conditions are extracted, a GMM for each of the machine conditions is built based on the corresponding extracted feature vectors, machine fault diagnosis can be accomplished through finding out the GMM whose posteriori probability for a given testing feature vector is the maximum of all. Experimental results based on the application on bearing fault diagnosis have shown that GMM can reliably diagnose not only the type of bearing faults, but also the degree of fault severity that are associated with incipient faults, moderate faults, and severe faults. Meanwhile, GMM has better diagnostic performance as compared to the multilayer perceptron neural networks.

Swallow segmentation with artificial neural networks and multi-sensor fusion

Swallow segmentation is a critical precursory step to the analysis of swallowing signal characteristics. In an effort to automatically segment swallows, we investigated artificial neural networks (ANN) with information from cervical dual-axis accelerometry, submental MMG, and nasal airflow. Our objectives were (1) to investigate the relationship between segmentation performance and the number of signal sources and (2) to identify the signals or signal combinations most useful for swallow segmentation. Signals were acquired from 17 healthy adults in both discrete and continuous swallowing tasks using five stimuli. Training and test feature vectors were constructed with variances from single or multiple signals, estimated within 200 ms moving windows with 50% overlap. Corresponding binary target labels (swallow or non-swallow) were derived by manual segmentation. A separate 3-layer ANN was trained for each participant–signal combination, and all possible signal combinations were investigated. As more signal sources were included, segmentation performance improved in terms of sensitivity, specificity, accuracy, and adjusted accuracy. The combination of all four signal sources achieved the highest mean accuracy and adjusted accuracy of 88.5% and 89.6%, respectively. A–P accelerometry proved to be the most discriminatory source, while the inclusion of MMG or nasal airflow resulted in the least performance improvement. These findings suggest that an ANN, multi-sensor fusion approach to segmentation is worthy of further investigation in swallowing studies.

Detection and Recognition of North Atlantic Right Whale Contact Calls in the Presence of Ambient Noise

In this paper, the problem of detecting and recognizing North Atlantic right whale (NARW), <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Eubalaena</i> <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">glacialis</i> , contact calls in the presence of ambient noise is considered. A proposed solution is based on a multistage, hypothesis-testing technique that involves the generalized likelihood ratio test (GLRT) detector, spectrogram testing, and feature vector testing algorithms. The main contributions of this paper are the inclusion of noise kernels for signals likely to produce false alarms and a second stage classification algorithm which extracts parameters from candidate contact calls and constructs a scaled squared error statistic for parameters which lie outside the range of expected calls. Closed-form representations of the algorithms are derived and realizable detection schemes are developed. Test results show that the proposed technique is able to detect approximately 80% of the contact calls detected by the human operator with about 26 false alarms per 24 h of observation. Testing data set included 44 227 right whale contact calls detected by eight human operators who performed visual and aural inspection of the data spectrogram. Data were collected in different periods from March 2001 to February 2007, in Cape Cod Bay, Great South Channel, and in the coastal waters of Georgia.

An efficient classifier to diagnose of schizophrenia based on the EEG signals

In this paper, electroencephalogram (EEG) signals of 13 schizophrenic patients and 18 age-matched control participants are analyzed with the objective of classifying the two groups. For each case, multi-channels (22 electrodes) scalp EEG is recorded. Several features including autoregressive (AR) model parameters, band power and fractal dimension are extracted from the recorded signals. Leave-one (participant)-out cross validation is used to have an accurate estimation for the separability of the two groups. Boosted version of Direct Linear Discriminant Analysis (BDLDA) is selected as an efficient classifier which applied on the extracted features. To have comparison, classifiers such as standard LDA, Adaboost, support vector machine (SVM), and fuzzy SVM (FSVM) are applied on the features. Results show that the BDLDA is more discriminative than others such that their classification rates are reported 87.51%, 85.36% and 85.41% for the BDLDA, LDA, Adaboost, respectively. Results of SVM and FSVM classifiers were lower than 50% accuracy because they are more sensitive to outlier instances. In order to determine robustness of the suggested classifier, noises with different amplitudes are added to the test feature vectors and robustness of the BDLDA was higher than the other compared classifiers.

Economic modeling of global test strategy I: Mathematical models

As electronic products become feature-rich, the amount of manufacturing testing required grows rapidly. This is further compounded because products made with components from several vendors, and in factories distributed all over the world, have similar tests repeated at various stages. While testing adds great value, it can also substantially increase manufacturing costs. In this paper we present a computational scheme that allows test strategists to investigate, and optimize, the nonintuitive trade-offs between test cost and product quality along the entire manufacturing chain. A method for constructing a mathematical model of very general test chains is presented that allows for the complete representation of realistic test and repair operations, down to the subtest level, including tests and repairs that have limited accuracy. A test-feature vector is introduced to account for test coverage. In addition, the notion of good and bad is introduced for every feature to allow the quantizing of quality. © 2007 Alcatel-Lucent.

Gait Recognition Using Radon Transform and Linear Discriminant Analysis

A new feature extraction process is proposed for gait representation and recognition. The new system is based on the Radon transform of binary silhouettes. For each gait sequence, the transformed silhouettes are used for the computation of a template. The set of all templates is subsequently subjected to linear discriminant analysis and subspace projection. In this manner, each gait sequence is described using a low-dimensional feature vector consisting of selected Radon template coefficients. Given a test feature vector, gait recognition and verification is achieved by appropriately comparing it to feature vectors in a reference gait database. By using the new system on the Gait Challenge database, very considerable improvements in recognition performance are seen in comparison to state-of-the-art methods for gait recognition.

ON THE DESIGN OF A TREE CLASSIFIER AND ITS APPLICATON TO SPEECH RECOGNITION

A new algorithm for constructing entropy reduction based decision tree classifier in two steps for large reference-class sets is proposed. The d-dimensional feature space is first mapped onto a line thus allowing a dynamic choice of features and then the resultant linear space is partitioned into two sections while minimizing the average system entropy. Classes of each section, again considered as a collection of references classes in a d-dimensional feature space, can be further split in a similar manner should the collection still be considered excessively large, thus forming a binary decision tree of nodes with overlapping members. The advantage of using such a classifier is that the need to match a test feature vector against all the references exhaustively is avoided. We demonstrate in this paper that discrete syllable recognition with dynamic programming equipped with such a classifier can reduce the recognition time by a factor of 40 to 100. The recognition speed is one third to one half of that using hidden Markov models (HMM), while the recognition rate is somewhat higher. The theory is considerably simpler than that of HMM but the decision tree can occupy a lot of memory space.

Fast search algorithm for VQ-based recognition of isolated words

This paper presents a fast search algorithm for vector quantisation (VQ)-based recognition of isolated words. It incorporates the property of high correlation between speech feature vectors of consecutive frames with the method of triangular inequality elimination to relieve the computational burden of vector-quantising the test feature vectors by full codebook search, and uses the extended partial distortion method to compress the incomplete matching computations of wildly mismatched words. Overall computational load can therefore be drastically reduced while the recognition performance of full search can be retained. Experimental results show that about 93% of multiplications and additions can be saved with a little increase of both comparisons and memory space.