• ETRI Journal
• /
• v.34 no.5
• /
• pp.791-794
• /
• 2012

We propose an efficient framework to realistically render 3D faces with a reduced set of points. First, a robust active appearance model is presented to detect facial features in the projected faces under different illumination conditions. Then, an adaptive simplification of 3D faces is proposed to reduce the number of points, yet preserve the detected facial features. Finally, the point model is rendered directly, without such additional processing as parameterization of skin texture. This fully automatic framework is very effective in rendering massive facial data on mobile devices.

• International Journal of Contents
• /
• v.8 no.4
• /
• pp.56-63
• /
• 2012

In recent decades, analyzing the activities of human brain achieved some accomplishments by using the functional Magnetic Resonance Imaging (fMRI) technique. fMRI data provide a sequence of three-dimensional images related to human brain's activity which can be used to detect instantaneous cognitive states by applying machine learning methods. In this paper, we propose a new approach for distinguishing human's cognitive states such as "observing a picture" versus "reading a sentence" and "reading an affirmative sentence" versus "reading a negative sentence". Since fMRI data are high dimensional (about 100,000 features in each sample), extremely sparse and noisy, feature selection is a very important step for increasing classification accuracy and reducing processing time. We used the Fisher Discriminant Ratio to select the most powerful discriminative features from some Regions of Interest (ROIs). The experimental results showed that our approach achieved the best performance compared to other feature extraction methods with the average accuracy approximately 95.83% for the first study and 99.5% for the second study.

• MALSORI
• /
• no.47
• /
• pp.97-107
• /
• 2003

Speaker change detection is to detect automatically a point of time at which speaker was replaced. Since feature parameters used for speaker change detection depend not only on speaker characteristics but also on phonetic characteristics, spoken contents included in the feature parameters inevitably causes performance degradation of speaker change detection. In this paper, to alleviate this problem, a method to normalize phonetic variations in speech feature parameters is proposed for emphasizing changes due to speaker characteristics. Experimental results show that the proposed method improves the performance of speaker change detection.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.28 no.2
• /
• pp.105-127
• /
• 2003

In this paper, four different data mining techniques, two neural networks and two statistical modeling techniques, are compared in terms of prediction accuracy in the context of bankruptcy prediction. In business setting, how to accurately detect the condition of a firm has been an important event in the literature. In neural networks, Backpropagation (BP) network and the Kohonen self-organizing feature map, are selected and compared each other while in statistical modeling techniques, discriminant analysis and logistic regression are also performed to provide performance benchmarks for the neural network experiment. The findings suggest that the BP network is a better choice among the data mining tools compared. This paper also identified some distinctive characteristics of Kohonen self-organizing feature map.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.95-101
• /
• 2018

In this paper, we propose an image-based synthesis method that can effectively represent the spark effect in fire simulation. We use the real flame image or animated image as inputs and perform the following steps : 1) extract feature vectors from the image, 2) calculate artificial buoyancy, and 3) generate and advect spark textures. We detect the edge from images and then calculate the feature vectors to calculate the buoyancy. In the next step, we compute the high-quality buoyancy vector field by integrating the two-dimensional feature vector and the fluid equation. Finally, the spark texture is advect by buoyancy field. As a result, our method is performed much faster than the previous approach and high-quality results can be obtained easily and stably.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.11-15
• /
• 2020

In a typical initial setup of a PCB component inspection system, operators should manually input various information such as category, position, and inspection area for each component to be inspected, thus causing much inconvenience and longer setup time. Although there are many deep learning based object detectors, RetinaNet is regarded as one of best object detectors currently available. In this paper, a method using an extended RetinaNet is proposed that automatically detects its component category and position for each component mounted on PCBs from a high-resolution color input image. We extended the basic RetinaNet feature pyramid network by adding a feature pyramid layer having higher spatial resolution to the basic feature pyramid. It was demonstrated by experiments that the extended RetinaNet can detect successfully very small components that could be missed by the basic RetinaNet. Using the proposed method could enable automatic generation of inspection areas, thus considerably reducing the setup time of PCB component inspection systems.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.60-67
• /
• 2020

Recently, artificial intelligence related technologies including machine learning are being applied to various fields, and the demand is also increasing. In particular, with the development of AR, VR, and MR technologies related to image processing, the utilization of computer vision based on deep learning has increased. The algorithms for object recognition and detection based on deep learning required for image processing are diversified and advanced. Accordingly, problems that were difficult to solve with the existing methodology were solved more simply and easily by using deep learning. This paper introduces various deep learning-based object recognition and extraction algorithms used to detect and recognize various objects in an image and analyzes the technologies that attract attention.

• Journal of the Korean Society of Safety
• /
• v.20 no.3 s.71
• /
• pp.64-70
• /
• 2005

In this literature, fault detection and classification of faulty induction motors are carried out through Z-index and frequency analysis. Above frequency analysis refer Fourier transformation and Wavelet transformation. Z-index is defined as the similar form of energy function, also the faulty and healthy conditions are classified through Z-index. For the detection and classification feature extraction for the fault detection of an induction motor is carried out using the information from stator current. Fourier and Wavelet transforms are applied to detect the characteristics under the healthy and various faulty conditions. We can obtain feature vectors from two transformations, and the results illustrate that the feature vectors are complementary each other.

• International Journal of Computer Science & Network Security
• /
• v.21 no.1
• /
• pp.1-5
• /
• 2021

Cyberbullying is a problem that is faced in many cultures. Due to their popularity and interactive nature, social media platforms have also been affected by cyberbullying. Social media users from Arab countries have also reported being a target of cyberbullying. Machine learning techniques have been a prominent approach used by scientists to detect and battle this phenomenon. In this paper, we compare different machine learning algorithms for their performance in cyberbullying detection based on a labeled dataset of Arabic YouTube comments. Three machine learning models are considered, namely: Multinomial Naïve Bayes (MNB), Complement Naïve Bayes (CNB), and Linear Regression (LR). In addition, we experiment with two feature extraction methods, namely: Count Vectorizer and Tfidf Vectorizer. Our results show that, using count vectroizer feature extraction, the Logistic Regression model can outperform both Multinomial and Complement Naïve Bayes models. However, when using Tfidf vectorizer feature extraction, Complement Naive Bayes model can outperform the other two models.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.2
• /
• pp.36-44
• /
• 2009

In this paper, we suggest a system of detecting a vehicle with lane violation, which can detect the vehicle with lane violation, by using the feature point tracking. The whole algorithm in the suggested system of detecting a vehicle with lane violation is composed of three stages such as feature extraction, register and tracking in feature for the tracking-targeted vehicle, and detecting a vehicle with lane violation. In the stage of feature extraction, the feature is extracted from the inputted image by sing the feature-extraction algorithm available for the real-time processing. The extracted features are again selected the racking-targeted feature. The registered feature is tracked by using NCC(normalized cross correlation). Finally, whether or not lane violation is finally detected by using information on the tracked features. As a result of experimenting the suggested system by using the acquired image in the section with a ban on intervention, the excellent performance was shown with 99.09% for positive recognition ratio and 0.9% for error ratio. The fast processing speed could be obtained in 34.48 frames per second available for real-time processing.