• 한국멀티미디어학회논문지
• /
• 제18권10호
• /
• pp.1149-1156
• /
• 2015

This paper proposes the hand contour detector which is robust to noises. Existing methods reduce noises by applying morphology to extracted edges, detect finger tips by using the center of hands, or exploit the intersection of curves from hand area candidates based on J-value segmentation(JSEG). However, these approaches are so vulnerable to noises that are prone to detect non-hand parts. We propose the noise tolerant hand contour detection method in which non-skin area noises are removed by applying skin area detection, contour detection, and a threshold value. By using the implemented system, we observed that the system was successfully able to detect hand contours.

• 제어로봇시스템학회논문지
• /
• 제20권6호
• /
• pp.593-598
• /
• 2014

Hand gesture recognition based on vision is a challenging task in human-robot interaction. The sign language of finger spelling alphabets has been tested as a kind of hand gesture. In this paper, we test hand gesture recognition by detecting the contour shape and orientation of hand with visual image. The method has three stages, the first stage of finding hand component separated from the background image, the second stage of extracting the contour feature over the hand component and the last stage of comparing the feature with the reference features in the database. Here, finger spelling alphabets are used to verify the performance of our system and our method shows good performance to discriminate finger alphabets.

• 전자공학회논문지CI
• /
• 제45권4호
• /
• pp.34-40
• /
• 2008

본 논문에서는 손 모양 인식을 이용한 비전기반의 모바일 로봇제어 시스템을 제안한다. 손 모양을 인식하기 위해서는 움직이는 카메라로부터 정확한 손의 경계선을 추출하고 추적하는 것이 필요하다. 이를 위해 본 논문에서는 초기 윤곽선 위치 및 경계에 강건하고, 빠른 물체를 정확히 추적할 수 있는 mean shift를 이용한 활성 윤곽선 모델(ACM) 추적 방법을 개발하였다. 제안된 시스템은 손 검출기, 손 추적기, 손 모양 인식기, 로봇 제어기 4가지 모듈로 구성된다. 손 검출기는 영상에서 피부색 영역으로 정확한 모양을 손으로 추출한 이후 활성 윤곽선 모델(ACM) 과 mean shift를 사용하여 손 영역을 정확히 추적한다. 마지막으로 Hue 모멘트에 이용하여 손의 형태를 인식한다. 제안된 시스템의 적합성을 평가하기 위하여 2족 보행로봇 RCB-1에서 실험이 수행되었다. 실험 결과는 제안된 시스템의 효율성을 증명하였다.

• 한국정보통신학회논문지
• /
• 제21권7호
• /
• pp.1349-1354
• /
• 2017

본 논문에서는 임베디드 시스템에 활용할 수 있는 고속의 손동작 인식 알고리즘을 제안한다. 기존의 손동작 인식 알고리즘은 손의 윤곽선을 구성하는 모든 점을 추출하는 윤곽선 추적 과정의 계산복잡도가 높기 때문에 임베디드 시스템, 모바일 디바이스와 같은 저성능의 시스템에서의 활용에 어려움이 있었다. 제안하는 알고리즘은 윤곽선 추적 알고리즘을 사용하는 대신 동심원 추적을 응용하여 추상화된 손가락의 윤곽선을 추정한 다음 특징을 추출하여 손동작을 분류한다. 제안된 알고리즘은 평균 인식률은 95%이고 평균 수행시간은 1.29ms로서 기존의 윤곽선 추적 방식을 사용하는 알고리즘에 비해 최대 44%의 성능향상을 보였고 임베디드 시스템, 모바일 디바이스와 같은 저성능의 시스템에서의 활용가능성을 확인하였다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제17권3호
• /
• pp.473-482
• /
• 2017

Hand gesture recognition is regarded as new Human Computer Interaction (HCI) technologies for the next generation of mobile devices. Previous hand gesture implementation requires a large memory and computation power for hand segmentation, which fails to give real-time interaction with mobile devices to users. Therefore, in this paper, we presents a low latency and memory-efficient hand segmentation architecture for natural hand gesture recognition. To obtain both high memory-efficiency and low latency, we propose a streaming hand contour tracing unit and a fast contour filling unit. As a result, it achieves 7.14 ms latency with only 34.8 KB on-chip memory, which are 1.65 times less latency and 1.68 times less on-chip memory, respectively, compare to the best-in-class.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.585-588
• /
• 2008

본 논문은 손동작 인식을 위한 형태 유사도 측정 방법을 제안한다. 이를 위해 손 영역 획득과 유사도 측정 단계로 나눈다. 손 영역 획득은 YCbCr 칼라 공간을 이용하여 손 영역을 추출하며, filter와 Histogram분석을 통하여 노이즈를 제거한다. 그리고 손 형태 유사도 측정은 윤곽선을 추출한 후 인접 간선들 사이의 거리와 각도 관계로 TSR을 적용하여 손동작의 유사성을 측정하였다. 파악된 특징점으로부터 형태 유사도 값을 측정한 후, 이를 손동작을 인식하는데 활용한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권5호
• /
• pp.2363-2374
• /
• 2016

Communication occurs through verbal elements, which usually involve language, as well as non-verbal elements such as facial expressions, eye contact, and gestures. In particular, among these non-verbal elements, gestures are symbolic representations of physical, vocal, and emotional behaviors. This means that gestures can be signals toward a target or expressions of internal psychological processes, rather than simply movements of the body or hands. Moreover, gestures with such properties have been the focus of much research for a new interface in the NUI/NUX field. In this paper, we propose a method for recognizing the number of fingers and detecting the hand region based on the depth information and geometric features of the hand for application to an NUI/NUX. The hand region is detected by using depth information provided by the Kinect system, and the number of fingers is identified by comparing the distance between the contour and the center of the hand region. The contour is detected using the Suzuki85 algorithm, and the number of fingers is calculated by detecting the finger tips in a location at the maximum distance to compare the distances between three consecutive dots in the contour and the center point of the hand. The average recognition rate for the number of fingers is 98.6%, and the execution time is 0.065 ms for the algorithm used in the proposed method. Although this method is fast and its complexity is low, it shows a higher recognition rate and faster recognition speed than other methods. As an application example of the proposed method, this paper explains a Secret Door that recognizes a password by recognizing the number of fingers held up by a user.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제3권5호
• /
• pp.195-200
• /
• 2014

능동 개체 윤곽 추출의 대표적인 방법은 스네이크(Snake)와 레벨 셋(Level Set) 기술이다. 일반적으로 스네이크는 속도는 빠르나 개체 위상을 처리하는 데 제약이 있다. 그러나 레벨 셋은 속도는 느리지만 개체 위상에 관계없이 잘 처리할 수 있는 장점이 있다. 본 논문에서는 빠르고 복잡한 위상을 처리하기 위해 두 방법의 장점을 이용한 알고리즘을 제안한다. 알고리즘은 2단계로 구성된다. 첫 번째 단계는 스네이크를 사용하여 빠르게 개체의 대략적인 윤곽을 추출한 후 레벨 셋을 두 번째 적용하여 복잡한 개체 윤곽을 정확하게 추출한다. 제안한 알고리즘은 다양한 위상을 갖는 5개의 이진영상 및 2개의 자연영상에 적용하여 속도 및 윤곽 추출이 개선된 것을 보여 준다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제16권5호
• /
• pp.557-563
• /
• 2016

Most of the existing hand detection methods rely on the contour shape of hand after skin color segmentation. Such contour shape based computations, however, are not only susceptible to noise and other skin color segments but also inherently sequential and difficult to efficiently parallelize. In this paper, we implement and accelerate our in-house distance image based approach using CPU-GPU heterogeneous computing. Using emerging CPU-GPU heterogeneous computing technology, we achieved 5.0 times speed-up for $320{\times}240$ images, and 17.5 times for $640{\times}480$ images and our experiment demonstrates that our proposed distance image based hand detection is robust and fast, reaching up to 97.32% palm detection rate, 80.4% of which have more than 3 fingers detected on commodity processors.

• Archives of Plastic Surgery
• /
• 제43권6호
• /
• pp.551-558
• /
• 2016

Background Fascial free flaps have been widely used for reconstruction of the hand because they are thin. However, studies reporting objective data regarding the advantages of this approach are lacking. Thus, we report our experience with such flaps. Methods Forty-five cases of fascial free flaps between November 2006 and March 2014 were reviewed. Nine cases involving reconstructed dorsal or lateral defects were included. Four anterolateral thigh fascial free flaps and 5 lateral arm fascial free flaps were examined. Maximal flap contour was assessed by measuring reconstructed tissue thickness at the central area from the surface of the skin to below the bone in a vertical manner using ultrasonography and X-ray data. Contralateral regions were examined in the same manner and a comparative analysis was performed. A questionnaire survey regarding aesthetic satisfaction was also administered. Results All reconstructed parts had a thicker contour than the contralateral side. The average relative percentage of reconstructed tissue thickness was found to be 152% using ultrasonography and 143% using X-ray imaging. According to the aesthetic satisfaction survey, the average rate of satisfaction for patients was 62%, and satisfaction with the flap contour was 72%. Conclusions Using a fascial free flap, the reconstructed tissue was approximately $1.5{\times}$ as thick as the contour of the normal side, which led to positive responses regarding aesthetic satisfaction.