• The Journal of the Acoustical Society of Korea
• /
• v.20 no.8
• /
• pp.12-18
• /
• 2001

In this paper, we propose the Recurrent Neural Predictive HMM (RNPHMM). The RNPHMM is the hybrid network of the recurrent neural network and HMM. The predictive recurrent neural network trained to predict the future vector based on several last feature vectors, and defined every state of HMM. This method uses the prediction value from the predictive recurrent neural network, which is dynamically changing due to the effects of the previous feature vectors instead of the stable average vectors. The models of the RNPHMM are Elman network prediction HMM and Jordan network prediction HMM. In the experiment, we compared the recognition abilities of the RNPHMM as we increased the state number, prediction order, and number of hidden nodes for the isolated digits. As a result of the experiments, Elman network prediction HMM and Jordan network prediction HMM have good recognition ability as 98.5% for test data, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.5
• /
• pp.1676-1689
• /
• 2014

At present, the Content-Based Image Retrieval (CBIR) system has become a hot research topic in the computer vision field. In the CBIR system, the accurate extractions of low-level features can reduce the gaps between high-level semantics and improve retrieval precision. This paper puts forward a new retrieval method aiming at the problems of high computational complexities and low precision of global feature extraction algorithms. The establishment of the new retrieval method is on the basis of the SIFT and Harris (APISH) algorithm, and the salient region of interest points (SRIP) algorithm to satisfy users' interests in the specific targets of images. In the first place, by using the IPDSH and SRIP algorithms, we tested stable interest points and found salient regions. The interest points in the salient region were named as salient interest points. Secondary, we extracted the pseudo-Zernike moments of the salient interest points' neighborhood as the feature vectors. Finally, we calculated the similarities between query and database images. Finally, We conducted this experiment based on the Caltech-101 database. By studying the experiment, the results have shown that this new retrieval method can decrease the interference of unstable interest points in the regions of non-interests and improve the ratios of accuracy and recall.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.2
• /
• pp.98-105
• /
• 2007

In this paper, we propose the method and system which determines the condition for safe and unsafe lane changing. To determine the condition, first, the system sets up the Region of Interest(ROI) on the neighboring lane. Second, a dangerous vehicle is extracted during the line changing. Third, the condition is determined to wm or not by calculating the moving direction, relative distance md relative velocity. To set up the ROI, the only one side lane is detected and the interested region is expanded. Using the coordinate transformation method, the accuracy of the ROI raised. To correctly extract the vehicle on the neighboring lane, the Adaptive Background Update method and Image Segmentation method which uses the feature of the travelling road are used. The object which is extracted by the dangerous vehicle is calculated the relative distance, the relative velocity and the moving average. And then in order to ring, the direction of the vehicle and the condition for safe and unsafe is determined. As minimizes the interested region and uses the feature of the travelling road, the computational quantity is reduced and the accuracy is raised and a stable result on a travelling road images which demands a high speed calculation is showed.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2E
• /
• pp.5-11
• /
• 1995

In this paper, a method using a neural network was applied for the purpose of urilizing spatial features. The adopted model of neural network the three-layered architecture, and the training algorithm is the back-propagation algorithm. Co-occurrence matrix which is generated from original imge was used for imput pattern to the neural network in order to tolerate variations of patterns like rotation of displacement. Co-occurrence matrix is explained in appendix. To evaluate this method, classification was executed with this method and texture features method over the city area and sand area, which cannot be separated with the conventional method mentioned aboved. In the results of this method and texture features proposed by Haralick the method using texture features was separation rate of 67%~89%. On the contrary, the method using neural network proposed in this research was stable and high separation rate of 80%~98%.

• Journal of the Korea Society for Simulation
• /
• v.20 no.2
• /
• pp.77-85
• /
• 2011

It is believed that the anomalies or diseases of human organs are identified by the analysis of the patterns. This paper proposes a new classification technique for the identification of cancer disease using the proteome patterns obtained from two-dimensional polyacrylamide gel electrophoresis(2-D PAGE). In the new classification method, three different classification methods such as support vector machine(SVM), multi-layer perceptron(MLP) and k-nearest neighbor(k-NN) are extended by multi-boosting method in an array of subclassifiers and the results of each subclassifier are merged by ensemble method. Genetic algorithm was applied to obtain optimal feature set in each subclassifier. We applied our method to empirical data set from cancer research and the method showed the better accuracy and more stable performance than single classifier.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.11
• /
• pp.2137-2142
• /
• 2016

The electrical current generated by heart creates not only electric potential but also a magnetic field. In this study, the signals obtained magnetocardiogram(MCG) using 61 channel superconducting quantum interference device(SQUID) system, and the clinical significance of various feature parameters has been developed MCG. Neural network algorithm was used to perform the classification of ischemic heart disease. The MCG signal was obtained to facilitate the extraction of parameters through a process of pre-processing. The data used to research the normal group 10 and ischemic heart disease group 10 with visible signs of stable angina patients. The available clinical indicators were extracted by characteristic point, characteristic interval parameter, and amplitude ratio parameter. The extracted parameters are determined to analysis the significance and clinical parameters were defined. It is possible to classify ischemic heart disease using the MCG feature parameters as a neural network input.

• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.535-546
• /
• 2019

Recently, there have been a lot of researches on the whole face replacement system, but it is not easy to obtain stable results due to various attitudes, angles and facial diversity. To produce a natural synthesis result when replacing the face shown in the video image, technologies such as face area detection, feature extraction, face alignment, face area segmentation, 3D attitude adjustment and facial transposition should all operate at a precise level. And each technology must be able to be interdependently combined. The results of our analysis show that the difficulty of implementing the technology and contribution to the system in facial replacement technology has increased in facial feature point extraction and facial alignment technology. On the other hand, the difficulty of the facial transposition technique and the three-dimensional posture adjustment technique were low, but showed the need for development. In this paper, we propose four facial replacement models such as 2-D Faceswap, OpenPose, Deekfake, and Cycle GAN, which are suitable for the Cox platform. These models have the following features; i.e. these models include a suitable model for front face pose image conversion, face pose image with active body movement, and face movement with right and left side by 15 degrees, Generative Adversarial Network.

• Journal of Convergence for Information Technology
• /
• v.10 no.12
• /
• pp.31-38
• /
• 2020

Radar signal identification of electronic warfare is a technology that recognizes the pulse repetition interval (PRI) from a set of pulse description words (PDWs) generated by the signal receiver. Conventionally batch processing is widely used in which a number of PDWs are collected as a unit and identifies PRI from the batch. In this paper, we propose a feature extraction algorithm based on the streaming process. This technique does not wait to form a batch. Whenever a PDW(Pulse Description Word) is generated from the signal receiver, the streaming process tries to form a cluster of PDWs, and makes the DTOA (Difference of Time of Arrival) histogram, finds out the frame PRI based on the concentration ratio, and decides the number of stagger stages. Experiments proved that the proposed algorithm derives stable recognition results as the cluster size increases.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.6
• /
• pp.1516-1529
• /
• 2023

In this paper, a motor gearbox fault-detection system based on a hierarchical flow-based model is proposed. The proposed system is used for the anomaly detection of a motion sound-based actuator module. The proposed flow-based model, which is a generative model, learns by directly modeling a data distribution function. As the objective function is the maximum likelihood value of the input data, the training is stable and simple to use for anomaly detection. The operation sound of a car's side-view mirror motor is converted into a Mel-spectrogram image, consisting of a folding signal and an unfolding signal, and used as training data in this experiment. The proposed system is composed of an encoder and a decoder. The data extracted from the layer of the pretrained feature extractor are used as the decoder input data in the encoder. This information is used in the decoder by performing an interlayer cross-scale convolution operation. The experimental results indicate that the context information of various dimensions extracted from the interlayer hierarchical data improves the defect detection accuracy. This paper is notable because it uses acoustic data and a normalizing flow model to detect outliers based on the features of experimental data.

• Journal of Ecology and Environment
• /
• v.30 no.4
• /
• pp.293-305
• /
• 2007

This study examined the distribution, population and vegetation structure of Corylopsis coreana in South Korea. C. coreana is distributed around the Suncheon area, Jeollanam-do, on the southern part of the Korean Peninsula, but the species is also found in Pocheon and Gangneung, on the central and central east parts of the peninsula. This discontinuous pattern of distribution is coupled with the unusual feature of only growing on northern exposed slopes. The mean density of C. coreana populations is 35 individuals per 100 $m^2$, ranging up to a maximum of 92 individuals per 100 $m^2$. Cut specimens sprouted a maximum of 38 stems per plant. Based on DCA analysis, the species' habitats was divided into three types by species composition and stratification structure. These types include: habitats affected strongly by human activities, valley and mantle communities which are affected relatively little by human activities, and stable forests. Populations affected by artificial intervention have actually flourished, while some populations in the stable forest system have declined. We conclude that the species, now endangered, should be maintained by means of specific external interventions such as cutting or removal of the canopy. To this end, further ecological data should be collected through monitoring and research to identify appropriate interventions to support threatened C. coreana populations.