• Smart Structures and Systems
• /
• v.13 no.3
• /
• pp.453-471
• /
• 2014

The fault diagnosis of rolling element bearings has drawn considerable research attention in recent years because these fundamental elements frequently suffer failures that could result in unexpected machine breakdowns. Artificial intelligence algorithms such as artificial neural networks (ANNs) and support vector machines (SVMs) have been widely investigated to identify various faults. However, as the useful life of a bearing deteriorates, identifying early bearing faults and evaluating their sizes of development are necessary for timely maintenance actions to prevent accidents. This study proposes a new two-layer structure consisting of support vector regression machines (SVRMs) to recognize bearing fault patterns and track the fault sizes. The statistical parameters used to track the fault evolutions are first extracted to condense original vibration signals into a few compact features. The extracted features are then used to train the proposed two-layer SVRMs structure. Once these parameters of the proposed two-layer SVRMs structure are determined, the features extracted from other vibration signals can be used to predict the unknown bearing health conditions. The effectiveness of the proposed method is validated by experimental datasets collected from a test rig. The results demonstrate that the proposed method is highly accurate in differentiating between fault patterns and determining their fault severities. Further, comparisons are performed to show that the proposed method is better than some existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.621-629
• /
• 2016

This paper presents a decision method of middle ear disease which is developed in children and adults. In the proposed method, features are extracted from the middle ear disease images and normal images using HoG (histogram of oriented gradient) descriptor and the extracted features are learned by SVM (support vector machine) classifier. To obtain an input vector into SVM, an input image is resized to a predefined size and then the resized image is partitioned into 16 blocks each of which is partitioned into 4 sub-blocks (namely cell). Finally, the feature vector with 576 components is given by using HoG with 9 bins and it is used as SVM learning and classification. Input images are classified by SVM classifier based on the model of learning features. Experimental results show that the proposed method yields the precision of over 90% in decision.

• Journal of Advanced Navigation Technology
• /
• v.14 no.2
• /
• pp.272-277
• /
• 2010

AR(Augmented Reality) needs medium between real and virtual, world, and recognition techniques are necessary to track an object continuously. Optical tracking using marker is mainly used, but it takes time and is inconvenient to attach marker onto the target objects. Therefore, many researchers try to develop markerless tracking techniques nowaday. In this paper, we extract features and 3D position from 3D objects and suggest realtime tracking based on these features and positions, which do not use just coplanar features and 2D position. We extract features using SURF, get rotation matrix and translation vector of 3D object using POSIT with these features and track the object in real time. If the extracted features are nor enough and it fail to track the object, then new features are extracted and re-matched to recover the tracking. Also, we get rotation in matrix and translation vector of 3D object using POSIT and track the object in real time.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.525-528
• /
• 2009

Recently, image classification has been an important task in various fields. Generally, the performance of image classification is not good without the adjustment of image features. Therefore, it is desired that the way of automatic feature extraction. In this paper, we propose an image classification method which adjusts image features automatically. We assume that texture features are useful in image classification tasks because natural images are composed of several types of texture. Thus, the classification accuracy rate is improved by using distribution of texture features. We obtain texture features by calculating image features from a current considering pixel and its neighborhood pixels. And we calculate image features from distribution of textures feature. Those image features are adjusted to image classification tasks using Genetic Algorithm. We apply proposed method to classifying images into "head" or "non-head" and "male" or "female".

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2005.11a
• /
• pp.178-181
• /
• 2005

Support vector machines is a powerful learning algorithm and attempt to separate belonging to two given sets in N-dimensional real space by a nonlinear surface, often only implicitly dened by a kernel function. We described breast tissue images analyses using texture features from Haar wavelet transformed images to classify breast lesion of ductal organ Benign, DCIS and CA. The approach for creating a classifier is composed of 2 steps: feature extraction and classification. Therefore, in the feature extraction step, we extracted texture features from wavelet transformed images with $10{\times}$ magnification. In the classification step, we created four classifiers from each image of extracted features using SVM(Support Vector Machines). In this study, we conclude that the best classifier in histological sections of breast tissue in the texture features from second-level wavelet transformed images used in Polynomial function.

• The Journal of Korea Robotics Society
• /
• v.3 no.4
• /
• pp.263-269
• /
• 2008

Visual odometry is a popular approach to estimating robot motion using a monocular or stereo camera. This paper proposes a novel visual odometry scheme using a stereo camera for robust estimation of a 6 DOF motion in the dynamic environment. The false results of feature matching and the uncertainty of depth information provided by the camera can generate the outliers which deteriorate the estimation. The outliers are removed by analyzing the magnitude histogram of the motion vector of the corresponding features and the RANSAC algorithm. The features extracted from a dynamic object such as a human also makes the motion estimation inaccurate. To eliminate the effect of a dynamic object, several candidates of dynamic objects are generated by clustering the 3D position of features and each candidate is checked based on the standard deviation of features on whether it is a real dynamic object or not. The accuracy and practicality of the proposed scheme are verified by several experiments and comparisons with both IMU and wheel-based odometry. It is shown that the proposed scheme works well when wheel slip occurs or dynamic objects exist.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.468-488
• /
• 2016

The paper proposes a novel framework for 3D face verification using dimensionality reduction based on highly distinctive local features in the presence of illumination and expression variations. The histograms of efficient local descriptors are used to represent distinctively the facial images. For this purpose, different local descriptors are evaluated, Local Binary Patterns (LBP), Three-Patch Local Binary Patterns (TPLBP), Four-Patch Local Binary Patterns (FPLBP), Binarized Statistical Image Features (BSIF) and Local Phase Quantization (LPQ). Furthermore, experiments on the combinations of the four local descriptors at feature level using simply histograms concatenation are provided. The performance of the proposed approach is evaluated with different dimensionality reduction algorithms: Principal Component Analysis (PCA), Orthogonal Locality Preserving Projection (OLPP) and the combined PCA+EFM (Enhanced Fisher linear discriminate Model). Finally, multi-class Support Vector Machine (SVM) is used as a classifier to carry out the verification between imposters and customers. The proposed method has been tested on CASIA-3D face database and the experimental results show that our method achieves a high verification performance.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.9 no.2
• /
• pp.99-104
• /
• 2009

The turn-to-turn short is one major fault of the motor faults of BLDC motors and can appear frequently. When the fault happens, the motor can be operated without breakdown, but it is necessary to maintain the motor for continuous working. In past research, several methods have been applied to detect winding faults. The representative approaches have been focusing on current signals, which can give important information to extract features and to detect faults. In this study, current sensors were installed to measure signals for fault detection of BLDC motors. In this study, the Park's vector method was used to extract the features and to isolate the faults from the current measured by sensors. Because this method can consider the three-phase current values, it is useful to detect features from one-phase and three-phase faults. After extracting two-dimensional features, the final feature was generated by using the two-dimensional values using the distance equation. The values were used in fuzzy similarity to isolate the faults. Fuzzy similarity is an available tool to diagnose the fault without model generation and the fault was converted to the percentage value that can be considered as possibility of the fault.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1326-1336
• /
• 2018

This paper proposes a numerical approach to predict the critical flashover voltages of air gaps under lightning impulses. For an air gap, the impulse voltage waveform features and electric field features are defined to characterize its energy storage status before the initiation of breakdown. These features are taken as the input parameters of the predictive model established by support vector machine (SVM). Given an applied voltage range, the golden section search method is used to compute the prediction results efficiently. This method was applied to predict the critical flashover voltages of rod-rod, rod-plane and sphere-plane gaps over a wide range of gap lengths and impulse voltage waveshapes. The predicted results coincide well with the experimental data, with the same trends and acceptable errors. The mean absolute percentage errors of 6 groups of test samples are within 4.6%, which demonstrates the validity and accuracy of the predictive model. This method provides an effectual way to obtain the critical flashover voltage and might be helpful to estimate the safe clearances of air gaps for insulation design.

• Journal of Korea Multimedia Society
• /
• v.16 no.7
• /
• pp.795-801
• /
• 2013

Recently, support vector machine has been widely used in various application fields due to its superiority of classification performance comparing with decision tree and neural network. Since support vector machine is basically designed for the binary classification problem, output coding method to analyze the classification result of multiclass binary classifier is used for the application of support vector machine into the multiclass problem. However, previous feature selection method for output coding based support vector machine found the features to improve the overall classification accuracy instead of improving each classification accuracy of each classifier. In this paper, we propose the novel feature selection method to find the features for maximizing the classification accuracy of each binary classifier in output coding based support vector machine. Experimental result showed that proposed method significantly improved the classification accuracy comparing with previous feature selection method.