Output Of Support Vector Machine Research Articles

Support vector machine–based facial-expression recognition method combining shape and appearance

Facial expression recognition can be widely used for various applications, such as emotion-based human-machine interaction, intelligent robot interfaces, face recognition robust to expression variation, etc. Previous studies have been classified as either shape- or appearance-based recognition. The shape-based method has the disadvantage that the individual variance of facial feature points exists irrespective of similar expressions, which can cause a reduction of the recognition accuracy. The appearance-based method has a limitation in that the textural information of the face is very sensitive to variations in illumination. To overcome these problems, a new facial-expression recognition method is proposed, which combines both shape and appearance information, based on the support vector machine (SVM). This research is novel in the following three ways as compared to previous works. First, the facial feature points are automatically detected by using an active appearance model. From these, the shape-based recognition is performed by using the ratios between the facial feature points based on the facial-action coding system. Second, the SVM, which is trained to recognize the same and different expression classes, is proposed to combine two matching scores obtained from the shape- and appearance-based recognitions. Finally, a single SVM is trained to discriminate four different expressions, such as neutral, a smile, anger, and a scream. By determining the expression of the input facial image whose SVM output is at a minimum, the accuracy of the expression recognition is much enhanced. The experimental results showed that the recognition accuracy of the proposed method was better than previous researches and other fusion methods.

Application of voice signal collection and analysis in traditional Chinese medicine syndrome differentiation of deficiency and excess

To provide more objective basis for traditional Chinese medicine (TCM) syndrome differentiation of deficiency and excess by collecting and analyzing voice signals and extracting characteristic parameters. All of 308 samples including 150 samples of qi-deficiency, 52 yin-deficiency, 55 excess and 51 normal were collected by "Voice Collecting System of TCM" and analyzed by wavelet packet transform (WPT) combined with approximate entropy (ApEn). The characteristic parameters with remarkable differences were chosen as input vectors for support vector machine (SVM) classifier to obtain classification results. Comparison between the normal and non-healthy showed that ApEn values of several nodes in 2 to 3 kHz, 3.5 to 4 kHz and 5 to 8 kHz frequencies were significantly different (P≤0.05); comparison between the deficiency and excess showed that ApEn values of several nodes in 0 to 2.5 kHz and 6 to 6.5 kHz frequencies were significantly different (P≤0.05); comparison between the qi-deficiency and yin-deficiency showed that ApEn values of node in 2 to 4 kHz frequency were significantly different(P≤0.05). The outputs of SVM showed that accuracies of samples in each group had good classification results by analyzing the ApEn values in different frequencies through WPT. The methods of voice collection and analysis used in the auscultation of TCM can provide objective basis for syndrome differentiation of deficiency and excess.

Fingerprint classification based on subclass analysis using multiple templates of support vector machines

Fingerprint classification reduces the searching time of an automated fingerprint identification system. Since fingerprints have properties of intra-class diversities and inter-class similarities, the ambiguous example causes a difficult problem in the fingerprint classification. In order to address the problem, we have analyzed fingerprints' subclasses with multiple decision templates. It clusters the soft outputs of support vector machines (SVMs) into several sub-classes using the self-organizing maps, and estimates a localized template for each sub-class. For an input fingerprint, the proposed method matches the output vector of SVMs to each template and finally categorizes the sample into the class of the most similar template. Experimental results on the FingerCode dataset demonstrate the effectiveness of the subclass-based approach compared with previous methods.

Effective functional mapping of fMRI data with support‐vector machines

There is a growing interest in using support vector machines (SVMs) to classify and analyze fMRI signals, leading to a wide variety of applications ranging from brain state decoding to functional mapping of spatially and temporally distributed brain activations. Studies so far have generated functional maps using the vector of weight values generated by the SVM classification process, or alternatively by mapping the correlation coefficient between the fMRI signal at each voxel and the brain state determined by the SVM. However, these approaches are limited as they do not incorporate both the information involved in the SVM prediction of a brain state, namely, the BOLD activation at voxels and the degree of involvement of different voxels as indicated by their weight values. An important implication of the above point is that two different datasets of BOLD signals, presumably obtained from two different experiments, can potentially produce two identical hyperplanes irrespective of their differences in data distribution. Yet, the two sets of signal inputs could correspond to different functional maps. With this consideration, we propose a new method called Effect Mapping that is generated as a product of the weight vector and a newly computed vector of mutual information between BOLD activations at each voxel and the SVM output. By applying this method on neuroimaging data of overt motor execution in nine healthy volunteers, we demonstrate higher decoding accuracy indicating the greater efficacy of this method.

Evaluation of Morphological Texture Features for Mangrove Forest Mapping and Species Discrimination Using Multispectral IKONOS Imagery

This letter aims to exploit morphological textures in discriminating three mangrove species and surrounding environment with multispectral IKONOS imagery in a study area on the Caribbean coast of Panama. Morphological texture features are utilized to distinguish red (Rhizophora mangle), white (Laguncularia racemosa), and black (Avicennia germinans) mangroves and rainforest regions. Meanwhile, two fusion methods are presented, i.e., vector stacking and support vector machine (SVM) output fusion, for integrating the hybrid spectral-textural features. For comparison purposes, the object-based analysis and the gray-level co-occurrence matrix (GLCM) textures are adopted. Results revealed that the morphological feature opening by reconstruction (OBR) followed by closing by reconstruction (CBR) and its dual operator CBR followed by OBR gave very promising accuracies for both mangrove discrimination (89.1% and 91.1%, respectively) and forest mapping (91.4% and 93.7%, respectively), compared with the object-based analysis (80.5% for mangrove discrimination and 82.9% for forest mapping) and the GLCM method (81.9% and 87.2%, respectively). With respect to the spectral-textural information fusion algorithms, experiments showed that the SVM output fusion could obtain an additional 2.0% accuracy improvement than the vector-stacking approach.

Using Stacked Generalization to Combine SVMs in Magnitude and Shape Feature Spaces for Classification of Hyperspectral Data

This paper proposes to improve the classification accuracy of hyperspectral data with support vector machines (SVMs) by using stacked generalization (stacking) as well as the complementary information of magnitude and shape feature spaces. Stacking is a method to combine multiple classifiers by learning a meta-level (or level-1) classifier from the outputs of base-level (or level-0) classifiers (estimated via cross-validation). In the processing of hyperspectral data, magnitude features are the radiance values at different sensor bands, whereas shape features are the differences in direction rather than the magnitude of the spectral signatures. In particular, the proposed method is as follows: (1) SVMs trained in magnitude and shape feature spaces are adopted as level-0 classifiers (termed as level-0 SVMs); (2) outputs (decision values) of the level-0 SVMs are used as inputs (termed as meta-level features) of level-1 classifier, since the decision values contain much more information than class labels; (3) level-1 classifier adopts SVMs (level-1 SVMs) trained in the meta-level feature space. In addition, we also discuss the possibility of reducing the number of level-0 SVMs by meta-level feature selection and present one simple solution. Experiments on a benchmark hyperspectral data set demonstrate that our method significantly outperforms the methods with the single feature space and other combining methods, namely, simple voting, absolute maximum decision value, and stacking with class labels.

Texture Classification and Shape Statistics Variational Approach for Segmentation of Left Ventricle Tagged MR Images

分割带标记线核磁共振(tagged MR)图像是左心室运动重建的前提.由于标记线的加载破坏了左心室的轮廓边缘和区域灰度一致性,再加上乳突肌的存在,使带标记线核磁共振图像的左心室内外轮廓分割变得相当困难.在变分框架下,将纹理分类信息与形状统计先验知识引入Mumford-Shah模型中,提出了一种改进的分割带标记线核磁共振图像的左心室内外轮廓的方法.该方法基于支持向量机对S滤波器组提取的纹理特征的分类结果,构造了一种新的图像能量表示;针对乳突肌及边缘断裂现象,引入形状统计先验信息来约束曲线的演化.因为分割过程利用了有监督学习策略,较好地克服了标记线对左心室区域灰度的影响,提高了分割精度.实验结果表明,该方法较以往方法具有更高的分割精度和更好的稳定性.;Segmentation of left ventricle tagged MR images is the basis of ventricular motion reconstruction. In left ventricle tagged MR images, the boundaries are often obscured or corrupted by the tag lines and region inhomogeneity as well as the existence of papillary muscles. These factors increase the difficulty of segmenting the inner and outer contour of left ventricle precisely. This paper introduces texture classification information and shape statistical knowledge into the Mumford-Shah model and presents an improved texture classification and shape statistics variational approach for the segmentation of inner and outer contour of left ventricle. It uses the output of support vector machine (SVM) classifier relying on S filter banks to construct a new region-based image energy term. This approach can overcome the influence of tag lines because it makes use of the supervised classification strategy. The introduction of shape statistics can improve the segmentation with broken boundaries. Segmentation results of an entire cardiac period on an identical image layer and a comparison of mean absolute distance analysis between contours generated by this approach and that generated by hand demonstrate that this method can achieve a higher segmentation precision and a better stability than other various approaches.

Structured prediction by joint kernel support estimation

Discriminative techniques, such as conditional random fields (CRFs) or structure aware maximum-margin techniques (maximum margin Markov networks (M3N), structured output support vector machines (S-SVM)), are state-of-the-art in the prediction of structured data. However, to achieve good results these techniques require complete and reliable ground truth, which is not always available in realistic problems. Furthermore, training either CRFs or margin-based techniques is computationally costly, because the runtime of current training methods depends not only on the size of the training set but also on properties of the output space to which the training samples are assigned. We propose an alternative model for structured output prediction, Joint Kernel Support Estimation (JKSE), which is rather generative in nature as it relies on estimating the joint probability density of samples and labels in the training set. This makes it tolerant against incomplete or incorrect labels and also opens the possibility of learning in situations where more than one output label can be considered correct. At the same time, we avoid typical problems of generative models as we do not attempt to learn the full joint probability distribution, but we model only its support in a joint reproducing kernel Hilbert space. As a consequence, JKSE training is possible by an adaption of the classical one-class SVM procedure. The resulting optimization problem is convex and efficiently solvable even with tens of thousands of training examples. A particular advantage of JKSE is that the training speed depends only on the size of the training set, and not on the total size of the label space. No inference step during training is required (as M3N and S-SVM would) nor do we have calculate a partition function (as CRFs do). Experiments on realistic data show that, for suitable kernel functions, our method works efficiently and robustly in situations that discriminative techniques have problems with or that are computationally infeasible for them.

A CFD assessment of classifications for hypersonic inlet start/unstart phenomena

AbstractInlet start/unstart detection is one of the most important issues of hypersonic inlets and is also the foundation of protection controls of scramjets. In ground and flight tests, it is inevitably to introduce the sensor noises to the measurement system. How to overcome or weaken the influence of the sensor noises and the outer disturbances is an important issue to the control system of the engine. To solve this problem, the 2D inner steady flow of hypersonic inlets was numerically simulated in different freestream conditions and backpressures, and two different inlet unstart phenomena were analysed. The membership function for hypersonic inlet start/unstart can be obtained by using probabilistic output support vector machine, and the algorithm of multiple classifiers fusion is introduced. The variations of the classification accuracy with the intensity of the sensor noises and the number of the classifier were discussed respectively. In conclusion, it is useful to introduce the algorithm of support vector machine and multiple classifiers fusion to overcome or weaken the influence of the sensor noises on the classification accuracy of hypersonic inlet start/unstart. The number of the practical fusion classifiers needs a tradeoff between the fusion classification accuracy and the complexity of the classification system.

An Asbestos Detection Method from Microscope Images Using Support Vector Random Field of Local Color Features

In this paper, an asbestos detection method from microscope images is proposed. The asbestos particles have different colors in two specific angles of the polarizing plate. Therefore, human examiners use the color information to detect asbestos. To detect the asbestos by computer, we develop the detector based on Support Vector Machine (SVM) of local color features. However, when it is applied to each pixel independently, there are many false positives and negatives because it does not use the relation with neighboring pixels. To take into consideration of the relation with neighboring pixels, Conditional Random Field (CRF) with SVM outputs is used. We confirm that the accuracy of asbestos detection is improved by using the relation with neighboring pixels.

Support Vector Machines for Automated Recognition of Obstructive Sleep Apnoea Syndrome from Electrocardiogram Recordings

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Obstructive sleep apnea syndrome (OSAS) is associated with cardiovascular morbidity as well as excessive daytime sleepiness and poor quality of life. In this study, we apply a machine learning technique [support vector machines (SVMs)] for automated recognition of OSAS types from their nocturnal ECG recordings. A total of 125 sets of nocturnal ECG recordings acquired from normal subjects (OSAS<formula formulatype="inline"><tex Notation="TeX">${-}$</tex> </formula>) and subjects with OSAS (OSAS<formula formulatype="inline"><tex Notation="TeX">${+}$</tex></formula>), each of approximately 8 h in duration, were analyzed. Features extracted from successive wavelet coefficient levels after wavelet decomposition of signals due to heart rate variability (HRV) from RR intervals and ECG-derived respiration (EDR) from R waves of QRS amplitudes were used as inputs to the SVMs to recognize OSAS<formula formulatype="inline"> <tex Notation="TeX">${+/-}$</tex></formula> subjects. Using leave-one-out technique, the maximum accuracy of classification for 83 training sets was found to be 100% for SVMs using a subset of selected combination of HRV and EDR features. Independent test results on 42 subjects showed that it correctly recognized 24 out of 26 OSAS <formula formulatype="inline"><tex Notation="TeX">${+}$</tex></formula> subjects and 15 out of 16 OSAS<formula formulatype="inline"> <tex Notation="TeX">${-}$</tex></formula> subjects (accuracy <formula formulatype="inline"><tex Notation="TeX">${=}$</tex></formula> 92.85%; Cohen's <formula formulatype="inline"><tex Notation="TeX">$\kappa$</tex></formula> value of 0.85). For estimating the relative severity of OSAS, the posterior probabilities of SVM outputs were calculated and compared with respective apnea/hypopnea index. These results suggest superior performance of SVMs in OSAS recognition supported by wavelet-based features of ECG. The results demonstrate considerable potential in applying SVMs in an ECG-based screening device that can aid a sleep specialist in the initial assessment of patients with suspected OSAS. </para>

Global Soft Decision Employing Support Vector Machine For Speech Enhancement

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this letter, we propose a novel speech enhancement technique based on global soft decision incorporating a support vector machine (SVM). Global soft decision in the proposed approach is performed employing the probabilistic outputs of the SVM rather than the conventional Bayes' rule. Actually, global speech absence probability (GSAP) is determined by the sigmoid function based on key parameters estimated by the model-trust minimization algorithm of the SVM output. Improved results are obtained in terms of speech quality measures for various types of noise and at different signal-to-noise ratio (SNR) levels when the proposed SVM is adopted in the global soft decision for speech enhancement. </para>

Support Vector Machines for Automated Recognition of Obstructive Sleep Apnea Syndrome From ECG Recordings

Obstructive sleep apnea syndrome (OSAS) is associated with cardiovascular morbidity as well as excessive daytime sleepiness and poor quality of life. In this study, we apply a machine learning technique [support vector machines (SVMs)] for automated recognition of OSAS types from their nocturnal ECG recordings. A total of 125 sets of nocturnal ECG recordings acquired from normal subjects (OSAS - ) and subjects with OSAS (OSAS +), each of approximately 8 h in duration, were analyzed. Features extracted from successive wavelet coefficient levels after wavelet decomposition of signals due to heart rate variability (HRV) from RR intervals and ECG-derived respiration (EDR) from R waves of QRS amplitudes were used as inputs to the SVMs to recognize OSAS +/- subjects. Using leave-one-out technique, the maximum accuracy of classification for 83 training sets was found to be 100% for SVMs using a subset of selected combination of HRV and EDR features. Independent test results on 42 subjects showed that it correctly recognized 24 out of 26 OSAS + subjects and 15 out of 16 OSAS - subjects (accuracy = 92.85%; Cohen's kappa value of 0.85). For estimating the relative severity of OSAS, the posterior probabilities of SVM outputs were calculated and compared with respective apnea/hypopnea index. These results suggest superior performance of SVMs in OSAS recognition supported by wavelet-based features of ECG. The results demonstrate considerable potential in applying SVMs in an ECG-based screening device that can aid a sleep specialist in the initial assessment of patients with suspected OSAS.

Accurate text localization in images based on SVM output scores

In this paper, we propose a new approach for accurate text localization in images based on SVM (support vector machine) output scores. In general, SVM output scores for the verification of text candidates provide a measure of the closeness to the text. Up to the present, most researchers used the score for verifying the text candidate region whether it is text or not. However, we use the output score for refining the initial localized text lines and selecting the best localization result from the different pyramid levels. By means of the proposed approach, we can obtain more accurate text localization results. Our method has three modules: (1) text candidate detection based on edge-CCA (connected component analysis), (2) text candidate verification based on the classifier fusion of N-gray (normalized gray intensity) and CGV (constant gradient variance), and (3) text line refinement based on the SVM output score, color distribution and prior geometric knowledge. By means of experiments on a large news database, we demonstrate that our method achieves impressive performance with respect to the accuracy, robustness and efficiency.

Overt foot movement detection in one single Laplacian EEG derivation

In this work one single Laplacian derivation and a full description of band power values in a broad frequency band are used to detect brisk foot movement execution in the ongoing EEG. Two support vector machines (SVM) are trained to detect the event-related desynchronization (ERD) during motor execution and the following beta rebound (event-related synchronization, ERS) independently. Their performance is measured through the simulation of an asynchronous brain switch. ERS (true positive rate = 0.74 ± 0.21) after motor execution is shown to be more stable than ERD (true positive rate = 0.21 ± 0.12). A novel combination of ERD and post-movement ERS is introduced. The SVM outputs are combined with a product rule to merge ERD and ERS detection. For this novel approach the average information transfer rate obtained was 11.19 ± 3.61 bits/min.

Protein Fold Recognition Based on Error Correcting Output Codes and SVM

A new approach based on the implementation of support vector machine (SVM) with the error correcting output codes (ECOC) is presented for recognition of multi-class protein folds. The experimental show that the proposed method can improve prediction accuracy by 4%-10% on two datasets containing 27 SCOP folds.

Albemarle starts new Texas catalyst plant

Support vector machine (SVM) was applied for land-cover characterization using MODIS time-series data. Classification performance was examined with respect to training sample size, sample variability, and landscape homogeneity (purity). The results were compared to two conventional nonparametric image classification algorithms: multilayer perceptron neural networks (NN) and classification and regression trees (CART). For 2001 MODIS time-series data, SVM generated overall accuracies ranging from 77% to 80% for training sample sizes from 20 to 800 pixels per class, compared to 67–76% and 62–73% for NN and CART, respectively. These results indicated that SVM’s had superior generalization capability, particularly with respect to small training sample sizes. There was also less variability of SVM performance when classification trials were repeated using different training sets. Additionally, classification accuracies were directly related to sample homogeneity/heterogeneity. The overall accuracies for the SVM algorithm were 91% (Kappa = 0.77) and 64% (Kappa = 0.34) for homogeneous and heterogeneous pixels, respectively. The inclusion of heterogeneous pixels in the training sample did not increase overall accuracies. Also, the SVM performance was examined for the classification of multiple year MODIS time-series data at annual intervals. Finally, using only the SVM output values, a method was developed to directly classify pixel purity. Approximately 65% of pixels within the Albemarle–Pamlico Basin study area were labeled as “functionally homogeneous” with an overall classification accuracy of 91% (Kappa = 0.79). The results indicated a high potential for regional scale operational land-cover characterization applications.

Wavelet-Based Feature Extraction for Support Vector Machines for Screening Balance Impairments in the Elderly

Trip related falls are a prevalent problem in the elderly. Early identification of at-risk gait can help prevent falls and injuries. The main aim of this study was to investigate the effectiveness of a wavelet based multiscale analysis of a gait variable [minimum foot clearance (MFC)] in comparison to MFC histogram plot analysis in extracting features for developing a model using support vector machines (SVMs) for screening of balance impairments in the elderly. MFC during walking on a treadmill was recorded on 13 healthy elderly and 10 elderly with a history of tripping falls. Features extracted from MFC histogram and then multiscale exponents between successive wavelet coefficient levels after wavelet decomposition of MFC series were used as inputs to the SVM to classify two gait patterns. The maximum accuracy of classification was found to be 100% for a SVM using a subset of selected wavelet based features, compared to 86.95% accuracy using statistical features. For estimating the relative risk of falls, the posterior probabilities of SVM outputs were calculated. These results suggest superior performance of SVM in the detection of balance impairments based on wavelet-based features and it could also be useful for evaluating for falls prevention intervention.

Prediction of tool breakage in face milling using support vector machine

A new approach is proposed using a support vector machine (SVM) to classify the feature of the cutting force signal for the prediction of tool breakage in face milling. The cutting force signal is compressed by averaging the cutting force signals per tooth to extract the feature of the cutting force signal due to tool breakage. With the SVM learning process, the output of SVM’s decision function can be utilized to identify a milling cutter with or without tool breakage. Experimental results are presented to verify the feasibility of this tool breakage prediction system in milling operations.

Feature selection via sensitivity analysis of SVM probabilistic outputs

Feature selection is an important aspect of solving data-mining and machine-learning problems. This paper proposes a feature-selection method for the Support Vector Machine (SVM) learning. Like most feature-selection methods, the proposed method ranks all features in decreasing order of importance so that more relevant features can be identified. It uses a novel criterion based on the probabilistic outputs of SVM. This criterion, termed Feature-based Sensitivity of Posterior Probabilities (FSPP), evaluates the importance of a specific feature by computing the aggregate value, over the feature space, of the absolute difference of the probabilistic outputs of SVM with and without the feature. The exact form of this criterion is not easily computable and approximation is needed. Four approximations, FSPP1-FSPP4, are proposed for this purpose. The first two approximations evaluate the criterion by randomly permuting the values of the feature among samples of the training data. They differ in their choices of the mapping function from standard SVM output to its probabilistic output: FSPP1 uses a simple threshold function while FSPP2 uses a sigmoid function. The second two directly approximate the criterion but differ in the smoothness assumptions of criterion with respect to the features. The performance of these approximations, used in an overall feature-selection scheme, is then evaluated on various artificial problems and real-world problems, including datasets from the recent Neural Information Processing Systems (NIPS) feature selection competition. FSPP1-3 show good performance consistently with FSPP2 being the best overall by a slight margin. The performance of FSPP2 is competitive with some of the best performing feature-selection methods in the literature on the datasets that we have tested. Its associated computations are modest and hence it is suitable as a feature-selection method for SVM applications.