Sungshin Kim

Wavelet analysis to fabric defects detection in weaving processes

The current procedure for the determination of fabric defects in the textile industry is performed by humans in the offline stage. The advantage of an online inspection system is not only defect detection and identification, but also quality improvement by a feedback control loop to adjust setpoints. This paper introduces a vision-based online fabric inspection methodology of woven textile fabrics. The proposed inspection system consists of hardware and software components. The hardware components consist of CCD array cameras, a frame grabber and appropriate illumination. The software routines capitalize upon vertical and horizontal scanning algorithms to reduce the 2-D image into a stream of 1-D data. The wavelet transform is used next to extract features that are characteristic of a particular defect. The signal to noise ratio (SNR) calculation based on the results of the wavelet transform is performed to measure any defects. The defect declaration is carried out by employing SNRs and scanning methods. Learning routines are called upon to optimize the wavelet coefficients. Test results from different types of defect and different styles of fabric demonstrate the effectiveness of the proposed inspection system.

e-prognosis and diagnosis for process management using data mining and artificial intelligence

In the past several decades, the huge amount of data was collected and processed by manufacturers to improve the quality and the productivity of products. Data collection mechanism as one of the process management system is an essential part in the manufacturing processes. Many researchers now devote substantial portions of their day to worrying about data handling that includes extracting information. But, the accumulated records in the real manufacturing processes are not effectively utilized to change operational conditions or remain unused condition. Therefore, the primary goal of this paper is to survey the existing KM techniques and apply the methods to two examples for e-prognosis and e-diagnosis purposes.

Indoor Localization for Mobile Robot using Extended Kalman Filter

This paper is presented an accurate localization scheme for mobile robots based on the fusion of ultrasonic satellite (U-SAT) with inertial navigation system (INS), i.e., sensor fusion. Our aim is to achieve enough accuracy less than 100 mm. The INS consist of a yaw gyro, two wheel-encoders. And the U-SAT consist of four transmitters, a receiver. Besides the localization method in this paper fuse these in an extended Kalman filter. The performance of the localization is verified by simulation and two actual data(straight, curve) gathered from about 0.5 m/s of driving actual driving data. localization methods used are general sensor fusion and sensor fusion through Kalman filter using data from INS. Through the simulation and actual data studies, the experiment show the effectiveness of the proposed method for autonomous mobile robots.

Fuzzy decision support system to the prediction of ozone concentrations

Artificial neural networks are used to model the interactions that occur between ozone concentration and environment data. A number of generic methods for analysis and modeling are investigated. Because the mechanism of the ozone concentration is highly complex, nonlinear, and nonstationary, the modeling methods of an ozone prediction system have many problems and the results of prediction are not of good performance so far, especially in the high-level ozone concentration. This paper introduces a modeling method of the ozone prediction system using neuro-fuzzy approaches and fuzzy clustering method. The dynamic polynomial neural network (DPNN) based upon a typical algorithm of GMDH (group method of data handling) is employed for data analysis, identification of nonlinear complex system, and prediction of the ozone concentration. The proposed prediction system is applied to the 19 areas in Seoul, Korea, and the final results are discussed in the senses of root mean-square error (RMSE) and R-square.

The Fault Diagnosis using Two-Steps Neural Networks for Nuclear Power Plants

Operating the nuclear power generations safely is not easy way because nuclear power generations are very complicated systems. In the main control room of the nuclear power generations, about 4000 numbers of alarms and monitoring devices are equipped to handle the signals corresponding to operating equipments. Thus, operators have to deal with massive information and to analyze the situation immediately. If they could not achieve these task, then they should make big problem in the power generations. Owing to too many variables, operators could be also in the uncontrolled situation. So in this paper, the fault diagnosis system is designed using 2-steps neural networks. This diagnosis method is based on the pattern of the principal variables which could represent the type and severity of faults.

High-Precision Positioning Control of Magnetic Levitation System

In this paper, we introduce two position control scheme; the lead-lag control and the sliding mode control for a stage system, which is levitated and driven by electric magnetic actuators. This consists of a levitating object (called platen) with 4 permanent magnetic linear synchronous motors in parallel. Each motor generates vertical force for suspension against gravity and propulsion force horizontally as well. This stage can generate six degrees of freedom motion by the vertical and horizontal forces. Dynamic equations of the stage system are derived simply. The sliding mode control algorithm is more effective than the lead-lag control algorithm to reduce effects from movements and disturbances of other axis

Extraction of quantitative and image information from flame images of steam boiler burners

Several types of detector, such as ultraviolet, infrared, visible light, differential pressure, flame rod, and others, are employed to detect fire flame in power generation plants. However, these flame detectors have some performance problems. This article describes the image-processing method of fire detection as well as neural network modeling. Nowadays, the image-processing technique is broadly applied in industrial fields. The neural network model has strong adaptability and learning capability, and is suitable for pattern classification. The Ulsan Steam Power Generation Plant in Korea was employed as the test field. If this technique can be implemented, boilers will be able to operate more economically and effectively.

An intelligent approach to integration of textile processes

This paper introduces a methodology to integrate effectively major plant processes with strong couplings between them. The proposed integration philosophy decides upon control setpoints for the individual processes by optimizing a global objective function which aims at improving process yield. A neuro-fuzzy model and a fuzzy objective function are employed to address the integration task. Such models and objective functions are defined and developed using experimental data or an operators experience. The objective is to maximize productivity and, at the same time, reduce defects in each of the subsequent operations. A textile plant is considered as a testbed and three processes-warping, slashing and weaving-are employed to illustrate the feasibility of the approach. The supervisory level of the control architecture is intended to continuously improve the control set points depending upon feedback information from the weave room, slasher operator and warping data.

An Emotion Recognition Technique using Speech Signals

In the field of development of human interface technology, the interactions between human and machine are important. The research on emotion recognition helps these interactions. This paper presents an algorithm for emotion recognition based on personalized speech signals. The proposed approach is trying to extract the characteristic of speech signal for emotion recognition using PLP (perceptual linear prediction) analysis. The PLP analysis technique was originally designed to suppress speaker dependent components in features used for automatic speech recognition, but later experiments demonstrated the efficiency of their use for speaker recognition tasks. So this paper proposed an algorithm that can easily evaluate the personal emotion from speech signals in real time using personalized emotion patterns that are made by PLP analysis. The experimental results show that the maximum recognition rate for the speaker dependant system is above 90%, whereas the average recognition rate is 75%. The proposed system has a simple structure and but efficient to be used in real time.

Modified Transformation and Evaluation for High Concentration Ozone Predictions

To reduce damage from high concentration ozone in the air, we have researched how to predict high concentration ozone before it occurs. High concentration ozone is a rare event and its reaction mechanism has nonlinearities and complexities. In this paper, we have tried to apply and consider as many methods as we could. We clustered the data using the fuzzy c-mean method and took a rejection sampling to fill in the missing and abnormal data. Next, correlations of the input component and output ozone concentration were calculated to transform more correlated components by modified log transformation. Then, we made the prediction models using Dynamic Polynomial Neural Networks. To select the optimal model, we adopted a minimum bias criterion. Finally, to evaluate suggested models, we compared the two models. One model was trained and tested by the transformed data and the other was not. We concluded that the modified transformation effected good to ideal performance In some evaluations. In particular, the data were related to seasonal characteristics or its variation trends.