• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.929-944
• /
• 2019

Many critical applications require accurate real-time human action recognition. However, there are many hurdles associated with capturing and pre-processing image data, calculating features, and classification because they consume significant resources for both storage and computation. To circumvent these hurdles, this paper presents a recognition machine learning (ML) based system model which uses reduced data structure features by projecting real 3D skeleton modality on virtual 2D space. The MMU VAAC dataset is used to test the proposed ML model. The results show a high accuracy rate of 97.88% which is only slightly lower than the accuracy when using the original 3D modality-based features but with a 75% reduction ratio from using RGB modality. These results motivate implementing the proposed recognition model on an embedded system platform in the future.

• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.113-121
• /
• 2020

Human Pose Estimation (HPE) which localizes the human body joints becomes a high potential for high-level applications in the field of computer vision. The main challenges of HPE in real-time are occlusion, illumination change and diversity of pose appearance. The single RGB image is fed into HPE framework in order to reduce the computation cost by using depth-independent device such as a common camera, webcam, or phone cam. However, HPE based on the single RGB is not able to solve the above challenges due to inherent characteristics of color or texture. On the other hand, depth information which is fed into HPE framework and detects the human body parts in 3D coordinates can be usefully used to solve the above challenges. However, the depth information-based HPE requires the depth-dependent device which has space constraint and is cost consuming. Especially, the result of depth information-based HPE is less reliable due to the requirement of pose initialization and less stabilization of frame tracking. Therefore, this paper proposes a new method of HPE which is robust in estimating self-occlusion. There are many human parts which can be occluded by other body parts. However, this paper focuses only on head self-occlusion. The new method is a combination of the RGB image-based HPE framework and the depth information-based HPE framework. We evaluated the performance of the proposed method by COCO Object Keypoint Similarity library. By taking an advantage of RGB image-based HPE method and depth information-based HPE method, our HPE method based on RGB-D achieved the mAP of 0.903 and mAR of 0.938. It proved that our method outperforms the RGB-based HPE and the depth-based HPE.

• 한국안전학회지
• /
• 제28권3호
• /
• pp.107-113
• /
• 2013

Problem solving and/or decision making process usually encountered in human living consists of a sequence of human behaviors based upon his/her knowledge. Thus, Rasmussen introduced Skill-Rule-Knowledge paradigm to countermeasure human errors that can occur in Nuclear Power Plants. Unfortunately however, it was not so easy as expected since objective evidence have not been obtainable with conventional research techniques. With the help of EEG band pawer ratio techniques, this study tried to get psycho-physiological symptoms of human errors, if any, while human beings perform knowledge-based behaviors such as simple arithmetic computations with different difficulty level. A set of simulated works was carried out with a computer station. Four kinds of arithmetic computation tasks were given to 10 health male under-graduate students on different day individually, and during the experiment, EEG and ECG was measured continuously for objective psycho-physiological analysis. According to the results, ${\alpha}$/(${\alpha}+{\beta}$) as well as ${\alpha}/{\beta}$ band power ratio were sensitive to task difficulty level which consistently decreased both. However, any one of them failed to reveal the influence of tasks with different difficulty level in the aspect of task duration time. On the contrary, Heart Rate Variability was more suggestive than expected. To make a conclusion, it can be said that band power of EEG waves will be helpful in not only assessment of work difficulty level but also assessment of workers' skill development if supported by cardiac function such as HRV.

• 디자인학연구
• /
• 제15권4호
• /
• pp.337-346
• /
• 2002

디자인 지식의 습득, 저장, 검색 및 응용과 같은 컴퓨터를 이용한 디자인 과정에 있어서, 창조적이며 디자인 요구에 적당한 결과물을 생산하는데 필요한 디자인 지식을 인지하고 습득하는 과정은 매우 중요하다 하겠다. 특히 인간의 인지능력과 유사한 기능을 같고 중요한 형태 디자인 지식을 습득하는 것은 필수적이다. 형태의 물리적인 속성에 의하여 인지되는 1차원적인 형태 지식이 아닌, 이들로부터 형성되는 2차원 또는 그 이상의 차원에서 인지되는 형태 디자인 지식을 인지해야만 한다. 지식의 인지 및 습득은 기억 장치에 저장되어 있는 지식과 인지되는 지식을 비교하여 동일하거나 유사한 경우 그 디자인 지식이 습득된다. 이때 1차원적인 디자인 지식은 형판 매칭과 속성 매칭에 의하여 그 유사성이 쉽게 인지되지만, 2차원 이상의 디자인 지식에 대해서는 인간은 쉽게 인지하나 컴퓨터를 이용한 인지에는 어려움이 많다. 본 연구는 컴퓨터에 이러한 능력을 부여하기 위하여 형태패턴 표현을 이용한 형태의 유사성을 판별하는 방법에 대하여 설명하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권12호
• /
• pp.3166-3179
• /
• 2013

A pre-processing system for adaptive noise removal is proposed based on the principle of identifying and filtering object regions and background regions. Human perception of images depends on bright, well-focused object regions; these regions can be treated with the best filters, while simpler filters can be applied to other regions to reduce overall computational complexity. In the proposed method, bright region segmentation is performed, followed by segmentation of object and background regions. Noise in dark, background, and object regions is then removed by the median, fast bilateral, and bilateral filters, respectively. Simulations show that the proposed algorithm is much faster than and performs nearly as well as the bilateral filter (which is considered a powerful noise removal algorithm); it reduces computation time by 19.4 % while reducing PSNR by only 1.57 % relative to bilateral filtering. Thus, the proposed algorithm remarkably reduces computation while maintaining accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권2호
• /
• pp.566-579
• /
• 2012

Ubiquitous computing called pervasive one is based on the thought of pervading ability of computation in daily life applications. In other words, it aims to include computation in devices such as electronic equipment and automobiles. This has led to disengagement of computers from desktop form. Accordingly, the notice in ubiquitous computing being taken of a world steeped in remote and wireless computer-based-services. Handheld and wearable programmed devices such as sense and control appliances are such devices. This advancement is rapidly moving domestic tasks and life from device-and-human communication to the device-and-device model. This model called Machine to Machine (M2M) has led to acceleration of developments in sciences such as nano-science, bio-science, and information science. As a result, M2M led to appearance of applications in various fields such as, environment monitoring, agricultural, health care, logistics, and business. Since it is envisaged that M2M communications will play a big role in the future in all wireless applications and will be emerged as a progressive linkage for next-generation communications, this paper aims to consider how much M2M architectures can realize ubiquitous computing in daily life applications. This is carried out after acquainting and initiating readers with M2M architectures and arguments for M2M. Some of the applications was not achievable before but are becoming viable owing to emergence of M2M communications.

• 한국정보통신학회논문지
• /
• 제12권7호
• /
• pp.1314-1320
• /
• 2008

최근 언제 어디서나 사람과 사물 같은 객체의 위치를 인식하고, 이를 기반으로 유용한 서비스를 제공하는 LBS(Location Based Service)가 대두되고 있다. LBS를 제공하기 위해 Cricket, Ubisense 등의 많은 연구가 진행되고 있지만, 이들은 노드들로만 구성된 네트워크를 이용하기 때문에 위치추적과 같은 복잡한 연산을 수행하는데 적합하지 않다. 본 연구는 위치 인식이 가능하고, 계산 능력이 높은 게이트웨이 노드를 구현하여 복잡한 연산을 가능하게 하였다.

• 한국정밀공학회지
• /
• 제16권7호
• /
• pp.109-118
• /
• 1999

A graphic software is one of the most important components of the vehicle simulator. To increase a visual reality of the simulator, the graphic software should require several technologies such as three-dimensional graphics, graphic modeling of the vehicle and the environment, drivers biomechanical models, and real-time data processing. This study presents a real time graphic driving simulator of a construction vehicle. The graphic simulator contains the three models of the construction vehicle, the human, and the environment, and employes a neural network approach to decrease an on-line dynamic computation. An excavator model is represented using an object-oriented paradigm and contains the detailed information about a real-size vehicle. The human model is introduced for objective visual evaluations of the developed excavator model. Since the environment model plays an important role in a real-time simulator, a block-based approach is implemented and a text format is utilized for easier construction of environment. The simulation results are illustrated in order to demonstrate the applicability of developed models and the neural network approach.

• 한국정밀공학회지
• /
• 제20권5호
• /
• pp.116-123
• /
• 2003

These days, the researches related with the emotional robots are actively investigated and in progress. And human language, expression, action etc. are merged in the emotional robot to understand the human emotion. However, there are so many sound sources and background noise around the robot, that the robots should be able to separate the mixture of these sound sources into the original sound sources, moreover to understand the meaning of voice of a specific person. Also they should be able to turn or move to the direction of a specific person to observe his expression or action effectively. Until now, the researches on the localization and separation of sound sources have been so theoretical and computative that real-time processing is hardly possible. In this reason for the practical emotional robot, fast computation should be realized by using simple principle. In this paper the methods for detecting the direction of sound sources by using the phase difference between peaks on spectrums, and the separating the sound sources by using fundamental frequency and its overtones of human voice, are proposed. Also by using these methods, it is shown that the effective and real-time localization and separation of sound sources in living room are possible.

• 대한조선학회논문집
• /
• 제47권1호
• /
• pp.58-66
• /
• 2010

Topology design optimization is a method to determine the optimal distribution of material that yields the minimal compliance of structures, satisfying the constraint of allowable material volume. The method is easy to implement and widely used so that it becomes a powerful design tool in various disciplines. In this paper, a large-scale topology design optimization method is developed using the efficient adjoint sensitivity and optimality criteria methods. Parallel computing technique is required for the efficient topology optimization as well as the precise analysis of large-scale problems. Parallelized finite element analysis consists of the domain decomposition and the boundary communication. The preconditioned conjugate gradient method is employed for the analysis of decomposed sub-domains. The developed parallel computing method in topology optimization is utilized to determine the optimal structural layout of human powered vessel.