• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.11
• /
• pp.54-64
• /
• 2002

This paper deals with the problem of synthesis of stable and optimal grasps with unknown objects by 3-finger hand. Previous robot grasp research has mainly analyzed with either unknown objects 2-dimensionally by vision sensor or known objects, such as cylindrical objects, 3-dimensionally. As extending the previous work, in this study we propose an algorithm to analyze grasp of unknown objects 3-dimensionally by using vision sensor. This is archived by two steps. The first step is to make a 3-dimensional geometrical model for unknown objects by using stereo matching. The second step is to find the optimal grasping points. In this step, we choose the 3-finger hand which has the characteristic of multi-finger hand and is easy to model. To find the optimal grasping points, genetic algorithm is employed and objective function minimizes the admissible force of finger tip applied to the objects. The algorithm is verified by computer simulation by which optimal grasping points of known objects with different angle are checked.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.10
• /
• pp.1946-1950
• /
• 2011

This paper deals with an input device without the touch. The existing touch screens have some problems such as the week durability by frequent contact and the high cost by complex hardware configuration. In this paper, a non-touch input device is proposed to overcome these problems. The proposed method uses a skin color generated by the HCbCr color model and a hand region obtained by the labeling technique. In Addition, the skeleton model is employed to improve the recognition performance of the hand motion. Finally, the experiment results show the applicability of the proposed method.

• Journal of Computational Design and Engineering
• /
• v.1 no.1
• /
• pp.1-12
• /
• 2014

In this paper, we propose a new method of reconstructing the hand models for individuals, which include the link structure models, the homologous skin surface models and the homologous tetrahedral mesh models in a reference posture. As for the link structure model, the local coordinate system related to each link consists of the joint rotation center and the axes of joint rotation, which can be estimated based on the trajectories of optimal markers on the relative skin surface region of the subject obtained from the motion capture system. The skin surface model is defined as a three-dimensional triangular mesh, obtained by deforming a template mesh so as to fit the landmark vertices to the relative marker positions obtained motion capture system. In this process, anatomical dimensions for the subject, manually measured by a caliper, are also used as the deformation constraints.

• Structural Engineering and Mechanics
• /
• v.29 no.1
• /
• pp.37-53
• /
• 2008

Plane steel portal frames, with pitched roof, exposed to fire, are examined. First, a determinate frame is analysed by hand. For flexible columns and shallow roof, snap-through occurs before plastic hinges mechanism is formed. An indeterminate frame with shorter columns and taller roof is also analysed by hand. Then, the same frame is simulated by a truss and a nonlinear static analysis is performed by use of a short computer program. The results of computer analysis by use of truss model are compared with those of analysis by hand and a satisfactory approximation between them is observed.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.563-565
• /
• 2023

Acupuncture and acupressure apply needles or pressure to anatomical points for therapeutic benefit. The over 350 mapped acupuncture points in the human body can each treat various conditions, but anatomical variations make precisely locating these acupoints difficult. We propose a computer vision technique using the real-time hand and face tracking capabilities of the MediaPipe framework to identify acupoint locations. Our model detects anatomical facial and hand landmarks, and then maps these to corresponding acupoint regions. In summary, our proposed model facilitates precise acupoint localization for self-treatment and enhances practitioners' abilities to deliver targeted acupuncture and acupressure therapies.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1994.04a
• /
• pp.596-605
• /
• 1994

An efficient measurement and evaluation system for hand tool tasks will provide a practical solution to the problem of designing and evaluating manual tool tasks in the workplace. Few studies on the biomechanical analysis of hand postures and tool handling tasks exist because of the lack of appropriate measurement techniques for hand force. A measurement system for the finger forces and joint angles for analysis of manual tool handling tasks was developed in this study. The measurement system consists of a force sensing glove made from twelve Force Sensitive Resistors and an angle-measuring glove (Cyberglove$\^$TM/, Virtual technologies) with eighteen joint angle sensors. A biomechanical model of the hand using the data from the measurement system was also developed. Systems of computerized procedures were implemented integrating the hand posture measurement system, biomechanical analysis system, and the task analysis system for manual tool handling tasks. The measurement system was useful in providing the hand force data needed for an existing task analysis system used in CTD risk evaluation. It is expected that the hand posture measurement developed in this study will provide an, efficient and cost-effective solution to task analysis of manual tool handling tasks. These tasks are becoming increasingly important areas of occupational health and safety of the country.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.384-386
• /
• 1998

The use of hand gesture provides an attractive alternative to cumbersome interface devices for human-computer interaction(HCI). Many methods hand gesture recognition using visual analysis have been proposed such as syntactical analysis, neural network(NN), Hidden Markov Model(HMM) and so on. In our research, a HMMs is proposed for alphabetical hand gesture recognition. In the preprocessing stage, the proposed approach consists of three different procedures for hand localization, hand tracking and gesture spotting. The hand location procedure detects the candidated regions on the basis of skin-color and motion in an image by using a color histogram matching and time-varying edge difference techniques. The hand tracking algorithm finds the centroid of a moving hand region, connect those centroids, and thus, produces a trajectory. The spotting a feature database, the proposed approach use the mesh feature code for codebook of HMM. In our experiments, 1300 alphabetical and 1300 untrained gestures are used for training and testing, respectively. Those experimental results demonstrate that the proposed approach yields a higher and satisfying recognition rate for the images with different sizes, shapes and skew angles.

• Journal of Industrial Technology
• /
• v.28 no.A
• /
• pp.155-161
• /
• 2008

Muscle force prediction in forward dynamic analysis of human motion depends many muscle parameters associated with muscle actuation. This research studies the effects of various parameters of Hill type muscle model using the simple hand raising motion. Motion analysis is carried out using motion capture system, and each muscle force is recorded for comparison with muscle model generated muscle force. Using Hill type muscle model, muscle force for generating the same hand rasing motion was setup adjusting 5 activation parameters. The test showed the importance of activation parameters on the accurate generation of muscle force.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.11-12
• /
• 2016

SVM (Support Vector machine) is powerful machine-learning method, and obtains better performance than traditional methods in the applications of muti-dimension nonlinear pattern classification. For the case of SVM model training and low efficiency in large samples, this paper proposes a combination of statistical parameters of the GMM-UBM (Universal Background Model) model. It is very effective to solve the problem of the large sample for the SVM training. The experiment is carried on four special dynamic hand gestures using the EPIC sensors. And the results show that the improved dynamic hand gesture recognition system has a high recognition rate up to 96.75%.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10b
• /
• pp.1774-1778
• /
• 1991

In this paper, we discuss the smooth hand-over of an object from a man to a robot and vice versa. In order for a robot to grasp an object or release a grasped object stably without using object model, as a man does, one of the basic approaches is the physiological method motivated by the study of human hands. So, we analyze human's grasping behavior by measuring grasp and friction forces simultaneously as a man grasps a experimental device which is designed for grasping or hand-over. Also, we investigate two methods that can predict when and bow fingers will slip upon a grasped object. And then, we propose a method of the hand-over of an object between a man and a robot by applying human's capability to a robot hand control.