• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.1077-1078
• /
• 2017

In this study, we implement a natural user interface/experience framework using multi-sensors: Microsoft Kinect, Leap Motion, and Myo Armband. The framework is designed for customers to use in various types of interactive applications. We integrate the functions of three sensors into an application and provide an interface for customers, who can use it to interact with a computer easily. The framework can track body information in real-time, and accurately recognize the motion of different body parts.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.1083-1085
• /
• 2017

개인별 건강관리에 대한 관심이 증가함에 따라 다양한 형태의 운동관리 프로그램이 개발되고 있다. 본 연구에서는 개인별 맞춤 트레이닝 관리를 위해 키넥트 센서를 활용한 셀프 운동 교정 프로그램을 개발하였다. 셀프 운동 교정 프로그램의 동작 과정은 다음과 같이 구성된다. 1)키넥트 센서를 활용하여 사용자의 운동 모습을 촬영 2) USB 어댑터를 이용하여 PC와 연동한 후 실시간으로 골격분석 및 좌표를 추출 및 각도를 계산 3)표준 자세의 데이터와 비교, 분석하여 잘못된 동작을 인식 4)잘못된 동작이 인식되면 음성지원을 통해 실시간 알림, 운동이 끝난 후 수집되는 운동 기록(횟수, 영상) 데이터를 데이터베이스에 저장하고 열람 및 관리할 수 있도록 함으로써 효율적으로 개인 운동교정이 가능하다.

• Journal of Korea Game Society
• /
• v.12 no.6
• /
• pp.5-14
• /
• 2012

The purpose of this paper is to demonstrate how three-dimensional gesture recognition technology, in children's traffic safety programs, can be an effective solution for instructing children in the safest ways to interact with traffic. In terms of traffic accidents, walking unaccompanied is the most dangerous traffic related activity for children. By using a three-dimensional serous game training program that implements gesture recognition, more accurate real life scenarios can be implemented in existing children's traffic training programs. The implementation of this technology will increase the possibility of changing the habits and attitudes of children, which in turn will lower the amount of walking related traffic accidents in children.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.541-544
• /
• 2012

In technology advances, The field of military and security surveillance system for a wide range of interest is required. Surveillance system offers a variety of services by the software that work with a remote access server. This paper shows the results of the implementation using three platforms as Mjpg-streamer for server, Arduino-multiwii for control system, OpenNI and OpenCV for image processing.

• Journal of Power System Engineering
• /
• v.17 no.5
• /
• pp.112-120
• /
• 2013

In this study, we propose an improved view-based navigation method for obstacle avoidance and evaluate the effectiveness of the method in real environments with real obstacles. The proposed method possesses the ability to estimate the position and rotation of a mobile robot, even if the mobile robot strays from a recording path for the purpose of avoiding obstacles. In order to achieve this, ego-motion estimation was incorporated into the existing view-based navigation system. The ego-motion is calculated from SURF points between a current view and a recorded view using a Kinect sensor. In conventional view-based navigation systems, it is difficult to generate alternate paths to avoid obstacles. The proposed method is anticipated to allow a mobile robot greater flexibility in path planning to avoid humans and objects expected in real environments. Based on experiments performed in an indoor environment using a mobile robot, we evaluated the measurement accuracy of the proposed method, and confirmed its feasibility for robot navigation in museums and shopping mall.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.07a
• /
• pp.151-152
• /
• 2017

본 논문에서는 키넥트 센서를 활용하여 노년층을 위한 헬스케어 어플리케이션을 설계하고 구현한다. 이 어플리케이션 메뉴는 전체 동작과 부분 동작으로 구성한다. 각 동작은 국민 건강증진센터에서 노년층에 장려하는 신체 각 부위별 스트레칭 10개 동작들을 구현한다. 신체의 각 부위는 온몸, 목, 무릎, 종아리로 분류한다. 전체 동작은 총 10개의 연속적인 동작들로 구성했으며 각 동작을 5초 이상 취할 경우 다음 동작으로 자동적으로 넘어가도록 구현한다. 각 동작이 진행되면서 동작 캐릭터 상태가 노년에서 청년으로 점점 젊어지게 함으로써 흥미를 느끼도록 한다. 부분 동작은 신체 어느 한 부분을 집중적으로 스트레칭 하고자 하는 사용자들에게 원하는 신체 부위(온몸, 목, 무릎, 종아리) 중 1개를 선택하여 집중적으로 스트레칭 할 수 있도록 구현한다. 전체적인 사운드는 어플리케이션의 분위기에 적합하도록 조용한 종류의 사운드로 구현하여 사용자로 하여금 편안한 감정을 느끼도록 구현한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.653-655
• /
• 2016

본 논문에서는 키넥트와 압력센서를 이용하여 수면자의 수면자세를 인식할 수 있는 수면자세 모니터링 시스템을 제안하였다. 기존 수면 모니터링 시스템은 장시간 착용해야 하는 불편함과 구속감으로 인해 수면의 질을 저하시킬 우려가 있다. 이러한 점을 해소하기 위해 압력센서와 키넥트 카메라를 이용하여 무구속 저비용의 효율적인 시스템을 구현하였고, 수면 매트형식으로 제작하여 그 유효성을 평가하였다. 본 연구에서 제안한 수면자세 모니터링 시스템은 실시간으로 수면자세를 감지하고 사용자의 수면시간 및 상태를 파악하여 건강한 수면습관을 들이는 방법을 권고할 수 있다. 향후에는 수집된 데이터를 이용하여 웰니스 및 헬스케어 모바일 응용 서비스로의 활용이 가능할 것이다.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1022-1034
• /
• 2023

In this study, the impact of varying lighting conditions on recognition and decision-making was considered. The luminosity approach was presented to increase gesture recognition performance under varied lighting. An efficient framework was proposed for sensor-based sign language gesture identification, including picture acquisition, preparing data, obtaining features, and recognition. The depth images were collected using multiple Microsoft Kinect devices, and data were acquired by varying resolutions to demonstrate the idea. A case study was designed to attain acceptable accuracy in gesture recognition under variant lighting. Using American Sign Language (ASL), the dataset was created and analyzed under various lighting conditions. In ASL-based images, significant feature points were selected using the scale-invariant feature transformation (SIFT). Finally, an artificial neural network (ANN) classified hand gestures using specified characteristics for validation. The suggested method was successful across a variety of illumination conditions and different image sizes. The total effectiveness of NN architecture was shown by the 97.6% recognition accuracy rate of 26 alphabets dataset with just a 2.4% error rate.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.39-46
• /
• 2021

In this paper, we propose a method for the motion recognition method required in the industrial field in mixed reality. In industrial sites, movements (grasping, lifting, and carrying) are required throughout the upper full-body, from trunk movements to arm movements. In this paper, we use a method composed of sensors and wearable devices that are not vision-based such as Kinect without using heavy motion capture equipment. We used two IMU sensors for the trunk and shoulder movement, and used Myo arm band for the arm movements. Real-time data coming from a total of 4 are fused to enable motion recognition for the entire upper body area. As an experimental method, a sensor was attached to the actual clothes, and objects were manipulated through synchronization. As a result, the method using the synchronization method has no errors in large and small operations. Finally, through the performance evaluation, the average result was 50 frames for single-handed operation on the HoloLens and 60 frames for both-handed operation.

• Journal of Korea Multimedia Society
• /
• v.16 no.6
• /
• pp.720-734
• /
• 2013

In this paper, we suggest an augmented reality authoring tool system that users can easily make augmented reality contents using hand gesture and marker-based interaction methods. The previous augmented reality authoring tools are focused on augmenting a virtual object and to interact with this kind of augmented reality contents, user used the method utilizing marker or sensor. We want to solve this limited interaction method problem by applying marker based interaction method and gesture interaction method using depth sensing camera, Kinect. In this suggested system, user can easily develop simple form of marker based augmented reality contents through interface. Also, not just providing fragmentary contents, this system provides methods that user can actively interact with augmented reality contents. This research provides two interaction methods, one is marker based method using two markers and the other is utilizing marker occlusion. In addition, by recognizing and tracking user's bare hand, this system provides gesture interaction method which can zoom-in, zoom-out, move and rotate object. From heuristic evaluation about authoring tool and compared usability about marker and gesture interaction, this study confirmed a positive result.