• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.482-486
• /
• 2010

Recently, gesture recognition methods based on dynamic time warping (DTW) have been actively investigated as more mobile devices have equipped the accelerometer. DTW has no additional training step since it uses given samples as the matching templates. However, it is difficult to apply the DTW on mobile environments because of its computational complexity of matching step where the input pattern has to be compared with every templates. In order to address the problem, this paper proposes a gesture recognition method based on DTW that uses localized subset of templates. Here, the k-means clustering algorithm is used to divide each class into subclasses in which the most centered sample in each subclass is employed as the localized template. It increases the recognition speed by reducing the number of matches while it minimizes the errors by preserving the diversities of the training patterns. Experimental results showed that the proposed method was about five times faster than the DTW with all training samples, and more stable than the randomly selected templates.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.243-244
• /
• 2017

최근 이미지 처리 및 인식 문제를 해결하는데 많이 사용되고 있는 CNN(Convolution Neural Network)를 이용하여 작은 dataset에서 Overfitting을 감소시키며 학습 할 수 있는 방법인 Dropout과 이미지를 왜곡하여 data를 늘리는 방법을 사용하여 보다 효율적으로 학습할 수 있는 방법을 연구 하였다. Batch별 처리속도를 기준으로 두 네트워크의 구조를 다르게 구현하여 비슷한 처리 시간을 수행하게 되도록 실험환경을 만들고 진행 하였다. Tensorflow로 네트워크를 구성하였고. Dataset은 Cifar_10을 사용 한다. 실험결과에 의하면 dropout의 경우 더 빨리 정확도가 향상되지만 이미지 왜곡을 사용하는 경우 저 높은 정확도로 수렴하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.6
• /
• pp.94-99
• /
• 1997

This paper describes a technique of probabilistic spectral subtraction which uses the knowledge of both noise and speech so as to reduce automatic speech recognition errors in noisy environments. Spectral subtraction method estimates a noise prototype in non-speech intervals and the spectrum of clean speech is obtained from the spectrum of noisy speech by subtracting this noise prototype. Thus noise can not be suppressed effectively using a single noise prototype in case the characteristics of the noise prototype are different from those of the noise contained in input noisy speech. To modify such a drawback, multiple noise prototypes are used in probabilistic subtraction method. In this paper, the probabilistic characteristics of noise and the knowledge of speech which is embedded in hidden Markov models trained in clean environments are used to suppress noise. Futhermore, dynamic feature parameters are considered as well as static feature parameters for effective noise suppression. The proposed method reduced error rates in the recognition of 50 Korean words. The recognition rate was 86.25% with the probabilistic subtraction, 72.75% without any noise suppression method and 80.25% with spectral subtraction at SNR(Signal-to-Noise Ratio) 10 dB.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.7
• /
• pp.1713-1723
• /
• 1999

Optical character recognition(OCR) might be the most plausible method in building database out of printed matters. This paper describes the points to be considered when one selects an OCR system in order to build database. Based on the considerations, we evaluated four commercial OCR systems, and chose one which shows the best recognition rate to build OCT-text database. The subject text, the KT-test collection, is a set of abstracts from proceedings of different printing quality, fonts, and formats. KT-test collection is also provided with typed text database. Recognition rate was calculated by comparing the recognition result with the typed text. No preprocessing such as learning and slant correction was applied to the recognition process in order to simulate a practical environment. The result shows 90.5% of character recognition rate over 970 abstracts. This recognition rate is still insufficient for practical use. The errors in OCR texts are different from those of manually typed texts. In this paper, we classify the errors in OCR texts for the further research.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.4
• /
• pp.838-845
• /
• 2004

After collecting the scanned images of practical identification licenses, it is attained to more accurate recognition of numbers in the identification licenses. As considering the process speed of the preprocess course for recognition, first, it is divided into eight equal parts of the identification license and then, removed the hologram of correspondent noises. It is run parallel template comparison method and teaming method for the number recognition and in order to extract a simple characteristics of the number the clustering method is used. Also, in case of misrecognized number because of external environment by run parallel with the thinning method, similar each numbers is sectioned by unique characteristics. From the results of number recognition, it is confirmed that the recognition rate of numbers is superior to other Studies.

• Journal of the Korean Chemical Society
• /
• v.67 no.6
• /
• pp.441-461
• /
• 2023

This study explore characteristics of teaching orientation and pck of science teachers in online-offline mixed learning environment. Data consisted of open-ended survey, semi-structured interview, class observation, field notes from 12 science teachers. We categorized teaching orientation considering both science education goals and science teaching·learning orientation. There were 8 different teaching orientations such as 'understanding science concepts-lecture centered' 'constructing science concepts-inquiry based' 'applying science concepts and inquiry-inquiry based' 'applying science concepts and inquiry-lectured centered' 'analyzing and judging science information-inquiry based' 'developing scientific attitude-inquiry based' 'developing scientific attitude-lecture centered' and 'developing perception of interrelationships among science, technology, and society-inquiry based'. Teachers with inquiry based teaching·learning orientation seemed to have knowledge of science curriculum specific to online learning environment for student inquiry. While teachers with 'understanding science concepts-lecture centered' teaching orientation appeared to have questioning strategy of checking student understanding and strategy of repeating a lecture, teachers with 'constructing science concepts-inquiry based' teaching orientation appeared to have knowledge of instructional strategies to perform online group activities targeting student construction of knowledge and to replace face-to-face group activities with virtual experiments and individual experiments. While teachers with 'understanding science concepts-lecture centered' teaching orientation did not show knowledge of student science learning, teachers with 'constructing science concepts-inquiry based' teaching orientation appeared to have knowledge of student difficulties in inquiry based learning.

• Journal of the Korean Chemical Society
• /
• v.47 no.1
• /
• pp.72-78
• /
• 2003

In this study, the effects of grouping by extraversion and introversion in POE learning applied to elementary science class on students' achievement, the perception of learning environment and the attitude toward science instruction were investigated. Ninety-five 5th graders were assigned to the control group and the experimental groups, and taught about acid and base for 9 class hours. In the experimental groups, the homogeneous small group was composed of four introverts or four extroverts, and the heterogeneous small group was composed of two introverts and two extroverts. Two-way ANCOVA results revealed that the homogeneous group performed better than the control group in the application subtest of the achievement test. Significant difference in learning difficulty was found between the heterogeneous group and the control group. In the attitude toward science class, significant interaction effect was found between the instruction and the extraversion/introversion.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.11
• /
• pp.193-201
• /
• 2014

This paper suggests a facial expression recognition system using face detection, face alignment, facial unit extraction, and training and testing algorithms based on AdaBoost classifiers. First, we find face region by a face detector. From the results, face alignment algorithm extracts feature points. The facial units are from a subset of action units generated by combining the obtained feature points. The facial units are generally more effective for smaller-sized databases, and are able to represent the facial expressions more efficiently and reduce the computation time, and hence can be applied to real-time scenarios. Experimental results in real scenarios showed that the proposed system has an excellent performance over 90% recognition rates.

• Proceedings of the Korean Society for the Gifted Conference
• /
• 2003.05a
• /
• pp.143-144
• /
• 2003

6차 교육과정부터는 과학 교육에서 학생들의 학습 목표를 과학 개념에 대한 이해 뿐 아니라 탐구 사고력 및 기능 향상도 중요한 목표로서 취급하고 있다(교육부, 1992). 그런데 학생의 탐구 사고력을 향상시킬 수 있는 과학교육이 제대로 이루어지지 못하고 있는 것이 현재의 실정이다. Schwab(1961)가 탐구 학습의 단계를 분류하면서 지적한 바 있듯이 탐구는 주어진 과제를 해결하는 고정적 탐구보다는 스스로 과제를 설정하고 이를 해결하려 하는 유동적 탐구가 더 바람직한 방향이며, 이러한 유형의 탐구를 통하여 학생들은 창의성의 신장과 함께 과학적탐구가 이루어지는 과정을 더 올바르게 이해할 수 있을 것으로 생각된다. 따라서 탐구의 상황을 학교 실험실만이 아닌 좀 더 생활 주변의 여러 가지 경험과 관련된 쪽으로 안내하는 것도 필요하다. 최근 활발한 컴퓨터의 보급과 인터넷 환경의 확대로 인하여 학생들이 이러한 환경에서 교사와 직접 동일한 시간, 동일한 장소에서 대면하지 않고도 의사소통하고 교수-학습이 이루어질 수 있는 기회가 사회적으로 가능해지고 있다. 이러한 원격교육은 교사의 안내에 따른 탐구 교수 형태의 개념 확인 및 검증 실험이 대부분인 전통적 과학학습 방법과 달리 학생 스스로 문제를 찾고 해결하려고 시도하는 것을 통하여 과학적 탐구 기능의 향상은 물론 과학적 개념의 획득, 과학, 사회, 기술에 대한 폭넓은 인식을 형성하는데 도움이 된다. 또한 인터넷 환경을 이용하면 학교 실험실 상황을 벗어나 학생들에게 다양한 탐구 활동 기회를 제공할 수 있고, 또한 그에 따른 의사 소통이 더 용이해질 수 있다. 이에 따라서 본 연구에서는 탐구 과정기술과 사고력을 중시하는 초등학교 과학과목의 특성을 고려하여 이에 적합한 인터넷 원격교수-학습을 위한 교수-학습 모형과 학생들의 과학적 탐구력과 사고력을 신장시킬 수 있는 멀티미디어 학습자료를 개발하고, 이를 실제적으로 적용할 수 있는 웹사이트를 개발, 현장에 적용하여 원격교수학습이 과학적 탐구력과 사고력에 미치는 효과에 대하여 조사하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.6
• /
• pp.565-570
• /
• 2013

The mirror neuron regions which are distributed in cortical area handled a functionality of intention recognition on the basis of imitative learning of an observed action which is acquired from visual-information of a goal-directed action. In this paper an automated intention recognition system is proposed by applying computational model of mirror neuron system to the human-robot interaction system. The computational model of mirror neuron system is designed by using dynamic neural networks which have model input which includes sequential feature vector set from the behaviors from the target object and actor and produce results as a form of motor data which can be used to perform the corresponding intentional action through the imitative learning and estimation procedures of the proposed computational model. The intention recognition framework is designed by a system which has a model input from KINECT sensor and has a model output by calculating the corresponding motor data within a virtual robot simulation environment on the basis of intention-related scenario with the limited experimental space and specified target object.