• 대한인간공학회지
• /
• 제19권1호
• /
• pp.91-100
• /
• 2000

In this paper the intent recognition system which recognizes the human's head movements as a simple non-verbal intent representation is presented. The system recognizes five basic intent representations. i.e., strong/weak affirmation. strong/weak negation, and ambiguity by image processing of nodding or shaking movements of head. The vision system for tracking the head movements is composed of CCD camera, image processing board and personal computer. The modified template matching method which replaces the reference image with the searched target image in the previous step is used for the robust tracking of the head movements. For the improvement of the processing speed, the searching is performed in the pyramid representation of the original image. By inspecting the variance of the head movement trajectories. we can recognizes the two basic intent representations - affirmation and negation. Also, by focusing the speed of the head movements, we can see the possibility which recognizes the strength of the intent representation.

• 한국음향학회지
• /
• 제18권3호
• /
• pp.10-16
• /
• 1999

본 연구에서는 소량의 음성 데이터만으로 적응화가 가능한 MAPE(최대사후확률추정)을 이용한 연속음성 인식시스템 개발에 대해 연구하였다. 음절단위 모델을 구축한 후 적응화 하고자 하는 화자의 데이터를 연결학습법과 Viterbi 알고리즘으로 음절단위의 추출을 자동화 한 후 MAPE로 적응화하였다. 자동차 제어문에 대해 화자 적응화한 경우의 인식률(O(n)DP인 경우)은 77.18%로 적응화 전의 결과보다 약 6%향상되었다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.1223-1226
• /
• 1998

The pre-emephasis filter as the conventional method emphasizes all components of high frequency that reflects the speaker characteristics. However this filter don't show the auditory characteristics of speaker's speech. In order to emphasize the perceptual characteristics, we propose the speaker recognition system that uses the perceptual weighting as the preprocessor because the Auditory characteristic of human is sensitive to the formant peaks. This filter has the characteristcs that both deemphasizes the low-formants and emphasizes the high formants. As a result of the proposed method, we improve the total recognition rate 1.7% better than the conventional method.

• 대한음성학회지:말소리
• /
• 제65호
• /
• pp.105-124
• /
• 2008

This paper presents the effects of different phone-like-unit (PLU) sets in order to propose an optimal PLU set for the performance improvement of Korean automatic speech recognition (ASR) systems. The examination of 9 currently used PLU sets indicates that most of them include a selection of allophones without any sufficient phonetic base. In this paper, a total of 34 PLU sets are designed based on Korean phonetic characteristics arid the effects of each PLU set are evaluated through experiments. The results show that the accuracy rate of each phone is influenced by different phonetic constraint(s) which determine(s) the PLU sets, and that an optimal PLU set can be anticipated through the phonetic analysis of the given speech data.

• 한국군사과학기술학회지
• /
• 제23권2호
• /
• pp.139-146
• /
• 2020

The camouflage pattern was difficult to distinguish from the surrounding background, so it was difficult to classify the object and the background image when the color image is used as the training data of deep-learning. In this paper, we proposed a red-green color blindness image transformation method using the principle that people of red-green blindness distinguish green color better than ordinary people. Experimental results show that the camouflage soldier's recognition performance improved by proposed a deep learning model of the ensemble technique using the imitated red-green-blind image data and the original color image data.

• 한국음향학회지
• /
• 제8권1호
• /
• pp.81-87
• /
• 1989

본 논문은 Markov 모델에 의한 효과적인 한국어 음소모델 작성방식과 인식에 대하여 기술한다. hidden Markov 모델은 음성신호 고유의 비정상성을 효과적으로 모델화할 수 있다. 본 논문에서는 음소의 일련의 변화하는 특성, 즉 천이-안정-천이의 변화를 나타내기 위하여 3상태 음소모델을 제안한다. 또한 음소길이가 인식성능에 영향을 미치는 중요한 요소임을 밝히고 길이를 고려한 3상태 hidden Markov 모델을 사용하여 인식률을 개선시킬 수 있음을 보였다.

• 전자공학회논문지C
• /
• 제35C권2호
• /
• pp.65-70
• /
• 1998

In this paper, we will to check and to correct mis-recognized word using Eojeol information. First, we divided into 16 classes that constituents in a Eojeol after we analyzed Korean statement into Eojeol units. Eojeol-Constituent state diagram constructed these constitutents, find the Left-Right Connectivity Information. As analogized the speech of connectivity information, reduced the number of cadidate words and restricted case of morphological analysis for mis-recognition Eojeol. Then, we improved correction speed uisng heuristic information as the adjacency information for Eojeol each other. In the correction phase, construct Reverse-Order Word Dictionary. Using this, we can trace word dictionary regardless of mis-recongnition word position. Its results show that improvement of recognition rate from 97.03% to 98.02% and check rate, reduction of chadidata words and morpholgical analysis cases.

• 음성과학
• /
• 제13권1호
• /
• pp.129-139
• /
• 2006

In this paper, we propose two effective energy feature normalization methods for robust speech recognition in noisy environments. In the first method, we estimate the noise energy and remove it from the noisy speech energy. In the second method, we propose a modified algorithm for the Log-energy Dynamic Range Normalization (ERN) method. In the ERN method, the log energy of the training data in a clean environment is transformed into the log energy in noisy environments. If the minimum log energy of the test data is outside of a pre-defined range, the log energy of the test data is also transformed. Since the ERN method has several weaknesses, we propose a modified transform scheme designed to reduce the residual mismatch that it produces. In the evaluation conducted on the Aurora2.0 database, we obtained a significant performance improvement.

• 대한음성학회지:말소리
• /
• 제53호
• /
• pp.103-118
• /
• 2005

The performance of speech recognition system is degraded by mismatch between training and test environments. Recently, Stereo-based Piecewise LInear Compensation for Environments (SPLICE) was introduced to overcome environmental mismatch using stereo data. In this paper, we propose several methods to improve the conventional SPLICE and evaluate them in the Aurora2 task. We generalize SPLICE to compensate for covariance matrix as well as mean vector in the feature space, and thereby yielding the error rate reduction of 48.93%. We also employ the weighted sum of correction vectors using posterior probabilities of all Gaussians, and the error rate reduction of 48.62% is achieved. With the combination of the above two methods, the error rate is reduced by 49.61% from the Aurora2 baseline system.

• 대한전자공학회논문지
• /
• 제27권6호
• /
• pp.954-962
• /
• 1990

Experties of recognizing an object despite of every possible occlusions among objects is difficult to be provided directly to a system. In this paper, we propose a method for inferring inherent shape-characteirstics of an object from training views provided. The method learns rules incrementally by alternating the rule induction process from limited number of training views and the rule verification process from the following taining views. The learned rules are represented using logical expressions to enhance the readability. Thr proposed method is tested by simulating occlusions on 2-dimensional objects to examine the learning process and to show improvement of recognition rate. Thr result shows that it can be applied to a practical system for 3-dimensional object recognition.