• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2010년도 춘계학술대회 논문집
• /
• pp.43-44
• /
• 2010

• 대한전자공학회논문지SP
• /
• 제42권6호
• /
• pp.167-176
• /
• 2005

본 논문은 음성인식에 쓰이는 음향모델의 모델링 방법 중 결정트리 상태공유 모델링(DTST)을 기반으로 출력 확률 분포의 혼합 가우시안 수를 줄여 모델을 최적화하는 방법을 제안한다. DTST는 음성학적 지식을 포함할 수 있는 질의어 집합과 유사도를 기반으로 한 결정 방법을 이용하는 것이다. 이때 상태들의 출력 확률 분포의 혼합 가우시안 수를 늘려 인식률을 증가시킬 수 있게 된다. 본 논문에서는 인식률이 최대가 되는 지점에서 혼합 가우시안들을 군집화 하여 그 수를 줄이고자 한다. 군집화 시에 필요한 거리 측정 방법은 유클리드(Euclidean)와 바타챠랴(Bhattacharyya) 방법을 이용하였고, 새로운 가우시안은 거리가 최소가 되는 두 가우시안으로부터 평균과 분산을 다시 계산하여 생성하였다. 증권상장 회사명(STOCKNAME) 1,680개의 단어 데이터베이스를 구성하여 실험한 결과 바타챠랴 방법은 $97.2\%$의 인식률을 유지하면서 전체 혼합 가우시안 수의 비율을 $1.0\%$로 감소시켰고, 유클리드 방법은 $96.9\%$의 인식률을 유지하면서 혼합 가우시안 수의 비율을 $1.0\%$로 감소시켜 모델을 최적화할 수 있었다.

• 한국산업정보학회논문지
• /
• 제22권5호
• /
• pp.25-30
• /
• 2017

본 논문에서는 초음파 센서를 이용한 저비용 장애물 인식 장치를 개발한다. 개발된 장애물 인식 장치는 시각 장애인들이 보행시 장애물을 인식하는 데 쓰일 수 있도록 한다. 장애물의 인식 여부는 내장된 모터의 진동을 통하여 시각 장애인들에게 인식이 가능하도록 한다. 뿐만 아니라 장애물의 거리에 따라 대상자에게 장애물 인식 유무에 대해 시간차를 두고 고지하여 대상자가 장애물에 대해 적절한 반응을 수행할 수 있도록 한다. 마이크로 컨트롤러 제어를 통한 펄스 신호를 이용하여 장애물에 대한 반사파를 통해 장애물의 존재 유무와 거리를 탐지하며 펄스 신호를 이용한 반사파 탐지는 반복 수행을 통해 인식률을 높인다. 개발된 장애물 인식 장치는 $30^{\circ}$ 내외의 전방 탐지각을 나타내고, 2cm-30cm의 탐지거리를 가지며 일상적인 보행 조건에서 동작이 가능함을 실험을 통해 입증하였다.

• 한국정보처리학회논문지
• /
• 제5권10호
• /
• pp.2600-2608
• /
• 1998

본 논문은 한글이나 한자처럼 문자 부류의 개수가 많은 경우에 효과적인 오프라인 필기 문자인식 알고리즘을 제안한다. 이 알고리즘은 간단하며 구현하기 쉬운 최소거리 분류법에 기반을 두고 있는데, 최소거리 분류법의 인식 성능을 향상시키기 위해 다단계 선인식(multi-stage pre-classification) 및 신경망을 이용한 후보문자 재정렬(candidate reordering)의 두 가지 전략이 첨가되었다. 제안된 알고리즘의 성능은 PE92 데이터베이스 상의 574 종의 한글 문자들에 대한 실험을 통해 입증하였는데, 인식률은 86%, 처리 속도는 초당 15자로서 기존의 연구 결과보다 우수함을 관측하였다.

• Journal of information and communication convergence engineering
• /
• 제11권2호
• /
• pp.124-131
• /
• 2013

Extraction and representation of postures and/or gestures from human activities in videos have been a focus of research in this area of action recognition. With various applications cropping up from different fields, this paper seeks to improve the performance of these action recognition machines by proposing a shape-based silhouette-edge descriptor for the human body. Information entropy, a method to measure the randomness of a sequence of symbols, is used to aid the selection of vital key postures from video frames. Morphological operations are applied to extract and stack edges to uniquely represent different actions shape-wise. To classify an action from a new input video, a Hausdorff distance measure is applied between the gallery representations and the query images formed from the proposed procedure. The method is tested on known public databases for its validation. An effective method of human action annotation and description has been effectively achieved.

• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.659-666
• /
• 2021

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2690-2701
• /
• 2013

Humans are able to recognize small number of people they know well by the way they walk. This ability represents basic motivation for using human gait as the means for biometric identification. Such biometric can be captured at public places from a distance without subject's collaboration, awareness or even consent. Although current approaches give encouraging results, we are still far from effective use in practical applications. In general, methods set various constraints to circumvent the influence factors like changes of view, walking speed, capture environment, clothing, footwear, object carrying, that have negative impact on recognition results. In this paper we investigate the influence of walking speed variation to different visual based gait recognition approaches and propose normalization based on geometric transformations, which mitigates its influence on recognition results. With the evaluation on MoBo gait dataset we demonstrate the benefits of using such normalization in combination with different types of gait recognition approaches.

• 전기학회논문지P
• /
• 제65권2호
• /
• pp.130-135
• /
• 2016

Gait recognition has advantage of non-contact type recognition. But It has disadvantage of low recognition rate when the pedestrian silhouette is changed due to bag or coat. In this paper, we proposed new method using combination of gait energy image feature and thermal face image feature. First, we extracted a face image which has optimal focusing value using human body rate and Tenengrad algorithm. Second step, we extracted features from gait energy image and thermal face image using linear discriminant analysis. Third, calculate euclidean distance between train data and test data, and optimize weights using genetic algorithm. Finally, we compute classification using nearest neighbor classification algorithm. So the proposed method shows a better result than the conventional method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권12호
• /
• pp.6063-6079
• /
• 2019

When the length of sampling data sequence is too large, the method of gesture recognition based on traditional Dynamic Time Warping (DTW) algorithm will lead to too long calculation time, and the accuracy of recognition result is not high.Support vector machine (SVM) has some shortcomings in precision, Edit Distance on Real Sequences(EDR) algorithm does not guarantee that noise suppression will not suppress effective data.A new method based on Improved Interpolation Dynamic Time Warping (IIDTW)algorithm is proposed to improve the efficiency of gesture recognition and the accuracy of gesture recognition. The results show that the computational efficiency of IIDTW algorithm is more than twice that of SVM-DTW algorithm, the error acceptance rate is FAR reduced by 0.01%, and the error rejection rate FRR is reduced by 0.5%.Gesture recognition based on IIDTW algorithm can achieve better recognition status. If it is applied to unlock mobile phone, it is expected to become a new generation of unlock mode.

• 전자공학회논문지B
• /
• 제28B권12호
• /
• pp.68-74
• /
• 1991

This paper proposes a HMM model based on fuzzy, as a method on the speech recognition of speaker-independent. In this recognition method, multi-observation sequences which give proper probabilities by fuzzy rule according to order of short distance from VQ codebook are obtained. Thereafter, the HMM model using this multi-observation sequences is generated, and in case of recognition, a word that has the most highest probability is selected as a recognized word. The vocabularies for recognition experiment are 146 DDD are names, and the feature parameter is 10S0thT LPC cepstrum coefficients. Besides the speech recognition experiments of proposed model, for comparison with it, we perform the experiments by DP, MSVQ and general HMM under same condition and data. Through the experiment results, it is proved that HMM model using fuzzy proposed in this paper is superior to DP method, MSVQ and general HMM model in recognition rate and computational time.