• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.23-34
• /
• 2004

In this paper we propose a feature point tracking algorithm using optical flow under non-prior taming active feature model (NPT-AFM). The proposed algorithm mainly focuses on analysis non-rigid objects[1], and provides real-time, robust tracking by NPT-AFM. NPT-AFM algorithm can be divided into two steps: (i) localization of an object-of-interest and (ii) prediction and correction of the object position by utilizing the inter-frame information. The localization step was realized by using a modified Shi-Tomasi's feature tracking algoriam[2] after motion-based segmentation. In the prediction-correction step, given feature points are continuously tracked by using optical flow method[3] and if a feature point cannot be properly tracked, temporal and spatial prediction schemes can be employed for that point until it becomes uncovered again. Feature points inside an object are estimated instead of its shape boundary, and are updated an element of the training set for AFH Experimental results, show that the proposed NPT-AFM-based algerian can robustly track non-rigid objects in real-time.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.2
• /
• pp.427-436
• /
• 2021

The purpose of this study is to implement a machine learning language model that learns literary texts. Literary texts have an important characteristic that pairs of question-and-answer are not frequently clearly distinguished. Also, literary texts consist of pronouns, figurative expressions, soliloquies, etc. They hinder the necessity of machine learning using literary texts by making it difficult to learn algorithms. Algorithms that learn literary texts can show more human-friendly interactions than algorithms that learn general sentences. For this goal, this paper proposes three text correction tasks that must be preceded in researches using literary texts for machine learning language model: pronoun processing, dialogue pair expansion, and data amplification. Learning data for artificial intelligence should have clear meanings to facilitate machine learning and to ensure high effectiveness. The introduction of special genres of texts such as literature into natural language processing research is expected not only to expand the learning area of machine learning, but to show a new language learning method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.361-365
• /
• 2007

위치정보는 유비쿼터스 컴퓨팅의 가장 중요한 항목 중 하나이다. 일반적인 위치 인식 시스템은 GPS가 대표적이지만, 실내에서 사용할 수 없고 건물내부와 같은 좁은 지역에서의 위치 인식이 어렵다는 단점이 있다. 특히 핸드폰, PDA와 같은 개인용 장비 에서는 더욱 정교한 위치 인식 기술이 필요한데, 무선랜을 기반으로 하는 위치 인식 기술은 그러한 목적을 달성하기에 적절하다. AP (Access Point)로부터 수집된 무선 신호의 세기는 모바일 기기의 위치를 측정하는데 필요한 지도로써 사용할 수 있지만, 건물의 벽, 사물, 사람 등과 같은 장애물의 간섭으로 변화가 심해 쉽게 사용할 수 없다. 본 논문에서는 이러한 문제점을 극복하기 위하여 신경망 모델을 이용한 무선랜 환경에서의 위치 인식 시스템을 제안한다. 아울러 신경망 학습에 사용될 학습데이터의 오차를 보정하고, 중복을 제거하기 위하여 칼만 필터를 사용하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.354-354
• /
• 2022

딥러닝(Deep Learning)은 일반적으로 인공신경망(Artificial Neural Network) 를 의미하는데, 이에 따른 결과는 데이터의 양, 변수, 학습모델의 학습횟수, 은닉층(Hidden Layer)의 개수 등 여러 요소로 인해 결정된다. 본 연구에서는 물리적 장기유출 모형인 SWAT의 결과를 참값으로 LSTM모형의 매개변수인 은닉층 갯수와 학습횟수등의 시나리오를 바탕으로 검보정을 수행하였으며, 최적의 목적함수를 갖는 매개변수를 도출하였다. 이를 이용하여 유량지속곡선을 도출한결과를 SWAT의 결과와 비교해본 결과 매우 높은 상관성을 도출하였으며 이를 통해 수자원분야에서 인공신경망의 활용 가능성을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.695-697
• /
• 2023

최근 몇 년 간 딥러닝 기반 모델의 규모와 복잡성이 증가하면서 강력하고, 높은 정확도가 확보되지만 많은 양의 계산 자원과 메모리가 필요하기 때문에 모바일 장치나 임베디드 시스템과 같은 리소스가 제한된 환경에서의 배포에 제약사항이 생긴다. 복잡한 딥러닝 모델의 배포 및 운영 시 요구되는 고성능 컴퓨터 자원의 문제점을 해결하고자 사전 학습된 대규모 모델로부터 가벼운 모델을 학습시키는 지식증류 기법이 제안되었다. 하지만 현대 딥러닝 기반 모델은 높은 정확도 대비 훈련 데이터에 과적합 되는 과잉 확신(overconfidence) 문제에 대한 대책이 필요하다. 본 논문은 효율적인 경량화를 위한 미리 학습된 모델의 과잉 확신을 방지하고자 초점 손실(focal loss)을 이용한 모델 보정 기법을 언급하며, 다양한 손실 함수 변형에 따라서 지식증류의 성능이 어떻게 변화하는지에 대해 탐구하고자 한다.

• Education of Primary School Mathematics
• /
• v.16 no.2
• /
• pp.163-182
• /
• 2013

The purpose of this study is an analysis on the 3rd graders' achievement in the elementary school. For this purpose, this study, first, analysed on the 3rd graders' achievement like the ratios of the achievement levels for whole students, sexual students, and regional students. Second, this study analysed the 3rd graders' assessment results like the total averages, averages for the contents area, sexual students, and regional students. Third, this study analysed students' special responses on the items.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.225-229
• /
• 2020

Recently, convolutional neural network (CNN) have been showed outstanding performance in the field of image recognition, image processing and computer vision, etc. In this paper, we propose a CNN-based image rotation correction algorithm as a solution to image rotation problem, which is one of the factors that reduce the recognition rate in image recognition system using CNN. In this paper, we trained our deep learning model with Leeds Sports Pose dataset to extract the information of the rotated angle, which is randomly set in specific range. The trained model is evaluated with mean absolute error (MAE) value over 100 test data images, and it is obtained 4.5951.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.663-670
• /
• 2019

Corrective thermal performance analysis is required for power plants' turbine cycles to determine the performance status of the cycle and improve the economic operation of the power plant. We developed a sectional classification method for the main feed-water flow to make precise corrections for the performance analysis based on the Performance Test Code (PTC) of the American Society of Mechanical Engineers (ASME). The method was developed for the estimation of the turbine cycle performance in a classified section. The classification is based on feature identification of the correlation status of the main feed-water flow measurements. We also developed predictive algorithms for the corrected main feed-water through a Kernel Regression (KR) model for each classified feature area. The method was compared with estimation using an Artificial Neural Network (ANN). The feature classification and predictive model provided more practical and reliable methods for the corrective thermal performance analysis of a turbine cycle.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.7
• /
• pp.83-90
• /
• 2010

In vocabulary recognition system has reduce recognition rate unrecognized error cause of similar phoneme recognition and due to provided inaccurate vocabulary. Input of inaccurate vocabulary by feature extraction case of recognition by appear result of unrecognized or similar phoneme recognized. Also can't feature extraction properly when phoneme recognition is similar phoneme recognition. In this paper propose vocabulary recognition post-process error correction system using phoneme likelihood based on phoneme feature. Phoneme likelihood is monophone training phoneme data by find out using MFCC and LPC feature extraction method. Similar phoneme is induced able to recognition of accurate phoneme due to inaccurate vocabulary provided unrecognized reduced error rate. Find out error correction using phoneme likelihood and confidence when vocabulary recognition perform error correction for error proved vocabulary. System performance comparison as a result of recognition improve represent MFCC 7.5%, LPC 5.3% by system using error pattern and system using semantic.

• Journal of Korea Water Resources Association
• /
• v.41 no.10
• /
• pp.1059-1065
• /
• 2008

Rainfall data is the most basic input data to analyze the hydrological phenomena and can be missing due to various reasons. In this research, a neural network based model to estimate missing rainfall data as approximate values was developed for 12 rainfall stations in the Soyang river basin to improve existing methods. This approach using neural network has shown to be useful in many applications to deal with complicated natural phenomena and displayed better results compared to the popular offline estimating methods, such as RDS(Reciprocal Distance Squared) method and AMM(Arithmetic Mean Method). Additionally, we proposed automated data reconciliation systems composed of a neural network learning processer to be capable of real-time reconciliation to transmit reliable hydrological data online.