• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.5
• /
• pp.487-496
• /
• 2000

In this paper, the Multi-FNN(Fuzzy-Neural Networks) model is identified and optimized using HCM(Hard C-Means) clustering method and genetic algorithms. The proposed Multi-FNN is based on Yamakawa's FNN and uses simplified inference as fuzzy inference method and error back propagation algorithm as learning rules. We use a HCM clustering and Genetic Algorithms(GAs) to identify both the structure and the parameters of a Multi-FNN model. Here, HCM clustering method, which is carried out for the process data preprocessing of system modeling, is utilized to determine the structure of Multi-FNN according to the divisions of input-output space using I/O process data. Also, the parameters of Multi-FNN model such as apexes of membership function, learning rates and momentum coefficients are adjusted using genetic algorithms. A aggregate performance index with a weighting factor is used to achieve a sound balance between approximation and generalization abilities of the model. The aggregate performance index stands for an aggregate objective function with a weighting factor to consider a mutual balance and dependency between approximation and predictive abilities. According to the selection and adjustment of a weighting factor of this aggregate abjective function which depends on the number of data and a certain degree of nonlinearity, we show that it is available and effective to design an optimal Multi-FNN model. To evaluate the performance of the proposed model, we use the time series data for gas furnace and the numerical data of nonlinear function.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1440-1447
• /
• 2019

Currently, the prevalence of autism spectrum disorders in children is reported to be higher and shows various types of disorders. In particular, they are having difficulty in communication due to communication impairment in the area of social communication and need to be improved through training. Thus, this study proposes a method of acquiring voice information through a microphone mounted on a robot designed through preliminary research and using this information to make intelligent motions. An ANN(Artificial Neural Network) was used to classify the speech data into robot motions, and we tried to improve the accuracy by combining the Recurrent Neural Network based on Convolutional Neural Network. The preprocessing of input speech data was analyzed using MFCC(Mel-Frequency Cepstral Coefficient), and the motion of the robot was estimated using various data normalization and neural network optimization techniques. In addition, the designed ANN showed a high accuracy by conducting an experiment comparing the accuracy with the existing architecture and the method of human intervention. In order to design robot motions with higher accuracy in the future and to apply them in the treatment and education environment of children with autism.

• Journal of IKEEE
• /
• v.11 no.4
• /
• pp.348-357
• /
• 2007

Due to the improved performance and cost of personal mobile devices and rapid progress of wireless communication technology, the number of users who utilize these devices is increasing. This trend requires various types of services be available to users. So far, there have been many solutions provided for the shortest path problem. But, technologies which can offer various recommendation services to user depending on user’s current location are focused on Euclidean spaces rather than road network. Thus, in this paper, we extend the previous work to satisfy this requirement on road network database. Our proposed scheme requires pre-computation for the efficient query processing. In the preprocessing step, we first partition the input road network into a fixed number of Voronoi polygons and then pre-compute routing information for each polygon. In the meantime, we select the number of Voronoi polygons in proposition to the scale of road network. Through this selection, the required size of pre-computation is linearly increasing to the size of road network. Using this pre-computated information, we can process queries more quickly. Through experiments, we have shown that our proposed scheme can achieve excellent performance in terms of scheduling time and the number of visited nodes.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.1
• /
• pp.97-107
• /
• 2002

In this paper, we propose the realization of on-line handwritten Chinese character recognition for mobile personal digital assistants (PDA). We focus on the development of an algorithm having a high recognition performance under the restriction that PDA requires small memory storage and less computational complexity in comparison with PC. Therefore, we use index matching method having computational advantage for fast recognition and we suggest a unit reconstruction method to minimize the memory size to store the character models and to accomodate the various changes in stroke order and stroke number of each person in handwriting Chinese characters. We set up standard model consisting of 1800 characters using a set of pre-defined units. Input data are measured by similarity among candidate characters selected on the basis of stroke numbers and region features after preprocessing and feature extracting. We consider 1800 Chinese characters adopted in the middle and high school in Korea. We take character sets of five person, written in printed style, irrespective of stroke ordering and stroke numbers. As experimental results, we obtained an average recognition time of 0.16 second per character and the successful recognition rate of 94.3% with MIPS R4000 CPU in PDA.

• The KIPS Transactions:PartC
• /
• v.8C no.6
• /
• pp.847-854
• /
• 2001

The Web continues to grow fast rate in both a large aclae volume of traffic and the size and complexity of Web sites. Along with growth, the complexity of tasks such as Web site design Web server design and of navigating simply through a Web site have increased. An important input to these design tasks is the analysis of how a web site is being used. The is paper proposes a Page logging System(PLS) identifying reliably user sessions required in Web mining system PLS consists of Page Logger acquiring all the page accesses of the user Log processor producing user session from these data, and statements to incorporate a call to page logger applet. Proposed PLS abbreviates several preprocessing tasks which spends a log of time and efforts that must be performed in Web mining systems. In particular, it simplifies the complexity of transaction identification phase through acquiring directly the amount of time a user stays on a page. Also PLS solves local cache hits and proxy IPs that create problems with identifying user sessions from Web sever log.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.161-170
• /
• 2019

Multi-product mass production small and medium enterprise factories have a wide variety of products and a large number of products, wasting manpower and expenses for inventory management. In addition, there is no way to check the status of inventory in real time, and it is suffering economic damage due to excess inventory and shortage of stock. There are many ways to build a real-time data collection environment, but most of them are difficult to afford for small and medium-sized companies. Therefore, smart factories of small and medium enterprises are faced with difficult reality and it is hard to find appropriate countermeasures. In this paper, we implemented the contents of extension of existing inventory management method through character extraction on label with barcode and QR code, which are widely adopted as current product management technology, and evaluated the effect. Technically, through preprocessing using OpenCV for automatic recognition and classification of stock labels and barcodes, which is a method for managing input and output of existing products through computer image processing, and OCR (Optical Character Recognition) function of Google vision API. And it is designed to recognize the barcode through Zbar. We propose a method to manage inventory by real-time image recognition through Raspberry Pi without using expensive equipment.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.23 no.1
• /
• pp.13-24
• /
• 2021

Tunnel boring machine (TBM) is widely used for tunnel excavation in hard rock and soft ground. In the perspective of TBM-based tunneling, one of the main challenges is to drive the machine optimally according to varying geological conditions, which could significantly lead to saving highly expensive costs by reducing the total operation time. Generally, drilling investigations are conducted to survey the geological ground before the TBM tunneling. However, it is difficult to provide the precise ground information over the whole tunnel path to operators because it acquires insufficient samples around the path sparsely and irregularly. To overcome this issue, in this study, we proposed a geological type classification system using the TBM operating data recorded in a 5 s sampling rate. We first categorized the various geological conditions (here, we limit to granite) as three geological types (i.e., rock, soil, and mixed type). Then, we applied the preprocessing methods including outlier rejection, normalization, and extracting input features, etc. We adopted a deep neural network (DNN), which has 6 hidden layers, to classify the geological types based on TBM operating data. We evaluated the classification system using the 10-fold cross-validation. Average classification accuracy presents the 75.4% (here, the total number of data were 388,639 samples). Our experimental results still need to improve accuracy but show that geology information classification technique based on TBM operating data could be utilized in the real environment to complement the sparse ground information.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.11-19
• /
• 2021

An abnormal object refers to a person, an object, or a mechanical device that performs abnormal and unusual behavior and needs observation or supervision. In order to detect this through artificial intelligence algorithm without continuous human intervention, a method of observing the specificity of temporal features using optical flow technique is widely used. In this study, an abnormal situation is identified by learning an algorithm that translates an input image frame to an optical flow image using a Generative Adversarial Network (GAN). In particular, we propose a technique that improves the pre-processing process to exclude unnecessary outliers and the post-processing process to increase the accuracy of identification in the test dataset after learning to improve the performance of the model's abnormal behavior identification. UCSD Pedestrian and UMN Unusual Crowd Activity were used as training datasets to detect abnormal behavior. For the proposed method, the frame-level AUC 0.9450 and EER 0.1317 were shown in the UCSD Ped2 dataset, which shows performance improvement compared to the models in the previous studies.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.227-241
• /
• 2022

Full waveform inversion (FWI) in the field of seismic data processing is an inversion technique that is used to estimate the velocity model of the subsurface for oil and gas exploration. Recently, deep learning (DL) technology has been increasingly used for seismic data processing, and its combination with FWI has attracted remarkable research efforts. For example, DL-based data processing techniques have been utilized for preprocessing input data for FWI, enabling the direct implementation of FWI through DL technology. DL-based FWI can be divided into the following methods: pure data-based, physics-based neural network, encoder-decoder, reparameterized FWI, and physics-informed neural network. In this review, we describe the theory and characteristics of the methods by systematizing them in the order of advancements. In the early days of DL-based FWI, the DL model predicted the velocity model by preparing a large training data set to adopt faithfully the basic principles of data science and apply a pure data-based prediction model. The current research trend is to supplement the shortcomings of the pure data-based approach using the loss function consisting of seismic data or physical information from the wave equation itself in deep neural networks. Based on these developments, DL-based FWI has evolved to not require a large amount of learning data, alleviating the cycle-skipping problem, which is an intrinsic limitation of FWI, and reducing computation times dramatically. The value of DL-based FWI is expected to increase continually in the processing of seismic data.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.