• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.241-241
• /
• 2012

Semiconductor industry has been taking the advantage of improvements in process technology in order to maintain reduced device geometries and stringent performance specifications. This results in semiconductor manufacturing processes became hundreds in sequence, it is continuously expected to be increased. This may in turn reduce the yield. With a large amount of investment at stake, this motivates tighter process control and fault diagnosis. The continuous improvement in semiconductor industry demands advancements in process control and monitoring to the same degree. Any fault in the process must be detected and classified with a high degree of precision, and it is desired to be diagnosed if possible. The detected abnormality in the system is then classified to locate the source of the variation. The performance of a fault detection system is directly reflected in the yield. Therefore a highly capable fault detection system is always desirable. In this research, time series modeling of the data from an etch equipment has been investigated for the ultimate purpose of fault diagnosis. The tool data consisted of number of different parameters each being recorded at fixed time points. As the data had been collected for a number of runs, it was not synchronized due to variable delays and offsets in data acquisition system and networks. The data was then synchronized using a variant of Dynamic Time Warping (DTW) algorithm. The AutoRegressive Integrated Moving Average (ARIMA) model was then applied on the synchronized data. The ARIMA model combines both the Autoregressive model and the Moving Average model to relate the present value of the time series to its past values. As the new values of parameters are received from the equipment, the model uses them and the previous ones to provide predictions of one step ahead for each parameter. The statistical comparison of these predictions with the actual values, gives us the each parameter's probability of fault, at each time point and (once a run gets finished) for each run. This work will be extended by applying a suitable probability generating function and combining the probabilities of different parameters using Dempster-Shafer Theory (DST). DST provides a way to combine evidence that is available from different sources and gives a joint degree of belief in a hypothesis. This will give us a combined belief of fault in the process with a high precision.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.189-198
• /
• 2024

Lipreading is one of the important parts of speech recognition, and several studies have been conducted to improve the performance of lipreading in lipreading systems for speech recognition. Recent studies have used method to modify the model architecture of lipreading system to improve recognition performance. Unlike previous research that improve recognition performance by modifying model architecture, we aim to improve recognition performance without any change in model architecture. In order to improve the recognition performance without modifying the model architecture, we refer to the cues used in human lipreading and set other regions such as chin and cheeks as regions of interest along with the lip region, which is the existing region of interest of lipreading systems, and compare the recognition rate of each region of interest to propose the highest performing region of interest In addition, assuming that the difference in normalization results caused by the difference in interpolation method during the process of normalizing the size of the region of interest affects the recognition performance, we interpolate the same region of interest using nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation, and compare the recognition rate of each interpolation method to propose the best performing interpolation method. Each region of interest was detected by training an object detection neural network, and dynamic time warping templates were generated by normalizing each region of interest, extracting and combining features, and mapping the dimensionality reduction of the combined features into a low-dimensional space. The recognition rate was evaluated by comparing the distance between the generated dynamic time warping templates and the data mapped to the low-dimensional space. In the comparison of regions of interest, the result of the region of interest containing only the lip region showed an average recognition rate of 97.36%, which is 3.44% higher than the average recognition rate of 93.92% in the previous study, and in the comparison of interpolation methods, the bilinear interpolation method performed 97.36%, which is 14.65% higher than the nearest neighbor interpolation method and 5.55% higher than the bicubic interpolation method. The code used in this study can be found a https://github.com/haraisi2/Lipreading-Systems.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.433-433
• /
• 2012

As the requirement in patterning geometry continuously shrinks down, the termination of etch process at the exact time became crucial for the success in nano patterning technology. By virtue of real-time optical emission spectroscopy (OES), etch end point detection (EPD) technique continuously develops; however, it also faced with difficulty in low open ratio etching, typically in self aligned contact (SAC) and one cylinder contact (OCS), because of very small amount of optical emission from by-product gas species in the bulk plasma glow discharge. In developing etching process, one may observe that coupon test is being performed. It consumes costs and time for preparing the patterned sample wafers every test in priority, so the coupon wafer test instead of the whole patterned wafer is beneficial for testing and developing etch process condition. We also can observe that etch open area is varied with the number of coupons on a dummy wafer. However, this can be a misleading in OES study. If the coupon wafer test are monitored using OES, we can conjecture the endpoint by experienced method, but considering by data, the materials for residual area by being etched open area are needed to consider. In this research, we compare and analysis the OES data for coupon wafer test results for monitoring about the conditions that the areas except the patterns on the coupon wafers for real-time process monitoring. In this research, we compared two cases, first one is etching the coupon wafers attached on the carrier wafer that is covered by the photoresist, and other case is etching the coupon wafers on the chuck. For comparing the emission intensity, we chose the four chemical species (SiF2, N2, CO, CN), and for comparing the etched profile, measured by scanning electron microscope (SEM). In addition, we adopted the Dynamic Time Warping (DTW) algorithm for analyzing the chose OES data patterns, and analysis the covariance and coefficient for statistical method. After the result, coupon wafers are over-etched for without carrier wafer groups, while with carrier wafer groups are under-etched. And the CN emission intensity has significant difference compare with OES raw data. Based on these results, it necessary to reasonable analysis of the OES data to adopt the pre-data processing and algorithms, and the result will influence the reliability for relation of coupon wafer test and whole wafer test.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.10
• /
• pp.723-730
• /
• 2013

This paper presents the design of an arm gesture recognition system using Kinect sensor. A variety of methods have been proposed for gesture recognition, ranging from the use of Dynamic Time Warping(DTW) to Hidden Markov Models(HMM). Our system learns a unique HMM corresponding to each arm gesture from a set of sequential skeleton data. Whenever the same gesture is performed, the trajectory of each joint captured by Kinect sensor may much differ from the previous, depending on the length and/or the orientation of the subject's arm. In order to obtain the robust performance independent of these conditions, the proposed system executes the feature transformation, in which the feature vectors of joint positions are transformed into those of angles between joints. To improve the computational efficiency for learning and using HMMs, our system also performs the k-means clustering to get one-dimensional integer sequences as inputs for discrete HMMs from high-dimensional real-number observation vectors. The dimension reduction and discretization can help our system use HMMs efficiently to recognize gestures in real-time environments. Finally, we demonstrate the recognition performance of our system through some experiments using two different datasets.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.149-160
• /
• 1997

This paper is concerned with the problem of speaker-adaptive speech synthes is method using a mapped codebook designed by fuzzy mapping on FLVQ (Fuzzy Learning Vector Quantization). The FLVQ is used to design both input and reference speaker's codebook. This algorithm is incorporated fuzzy membership function into the LVQ(learning vector quantization) networks. Unlike the LVQ algorithm, this algorithm minimizes the network output errors which are the differences of clas s membership target and actual membership values, and results to minimize the distances between training patterns and competing neurons. Speaker Adaptation in speech synthesis is performed as follow;input speaker's codebook is mapped a reference speaker's codebook in fuzzy concepts. The Fuzzy VQ mapping replaces a codevector preserving its fuzzy membership function. The codevector correspondence histogram is obtained by accumulating the vector correspondence along the DTW optimal path. We use the Fuzzy VQ mapping to design a mapped codebook. The mapped codebook is defined as a linear combination of reference speaker's vectors using each fuzzy histogram as a weighting function with membership values. In adaptive-speech synthesis stage, input speech is fuzzy vector-quantized by the mapped codcbook, and then FCM arithmetic is used to synthesize speech adapted to input speaker. The speaker adaption experiments are carried out using speech of males in their thirties as input speaker's speech, and a female in her twenties as reference speaker's speech. Speeches used in experiments are sentences /anyoung hasim nika/ and /good morning/. As a results of experiments, we obtained a synthesized speech adapted to input speaker.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.387-387
• /
• 2022

기존 홍수기시 댐 운영은 예측 강우와 실시간 관측 강우를 이용하여 댐 운영 모형을 수행하며, 예측 결과에 따라 의사결정 및 댐 운영을 실시하게 된다. 하지만 이 과정에서 반복적인 분석이 필요하며, 댐 운영 모형 수행자의 경험에 따라 예측 결과가 달라져서 반복작업에 대한 자동화, 모형 수행자에 따라 달라지지 않는 예측 결과의 일반화가 필요한 상황이다. 이에 댐 운영 모형에 AI 기법을 적용하여, 다양한 강우 상황에 따른 자동 예측 및 모형 결과의 일반화를 구현하고자 하였다. 이를 위해 수자원 분야에 적용된 국내외 129개 연구논문에서 사용된 딥러닝 기법의 활용성을 분석하였으며, 다양한 수자원 분야 AI 적용 사례 중에서 댐 운영 예측 모형에 적용한 사례는 없었지만 유사한 분야로는 장기 저수지 운영 예측과 댐 상·하류 수위, 유량 예측이 있었다. 수자원의 시계열 자료 활용을 위해서는 Long-Short Term Memory(LSTM) 기법의 적용 활용성이 높은 것으로 분석되었다. 댐 운영 모형에서 AI 적용은 2개 분야에서 진행하였다. 기존 강우관측소의 관측 강우를 활용하여 강우의 패턴분석을 수행하는 과정과, 강우에서 댐 유입량 산정시 매개변수 최적화 분야에 적용하였다. 강우 패턴분석에서는 유사한 표본끼리 묶음을 생성하는 K-means 클러스터링 알고리즘과 시계열 데이터의 유사도 분석 방법인 Dynamic Time Warping을 결합하여 적용하였다. 강우 패턴분석을 통해서 지점별로 월별, 태풍 및 장마기간에 가장 많이 관측되었던 강우 패턴을 제시하며, 이를 모형에서 직접적으로 활용할 수 있도록 구성하였다. 강우에서 댐 유입량을 산정시 활용되는 매개변수 최적화를 위해서는 3층의 Multi-Layer LSTM 기법과 경사하강법을 적용하였다. 매개변수 최적화에 적용되는 매개변수는 중권역별 8개이며, 매개변수 최적화 과정을 통해 산정되는 결과물은 실측값과 오차가 제일 적은 유량(유입량)이 된다. 댐 운영 모형에 AI 기법을 적용한 결과 기존 반복작업에 대한 자동화는 이뤘으며, 댐 운영에 따른 상·하류 제약사항 표출 기능을 추가하여 의사결정에 소요되는 시간도 많이 줄일 수 있었다. 하지만, 매개변수 최적화 부분에서 기존 댐운영 모형에 적용되어 있는 고전적인 매개변수 추정기법보다 추정시간이 오래 소요되며, 매개변수 추정결과의 일반화가 이뤄지지 않아 이 부분에 대한 추가적인 연구가 필요하다.