• Journal of Welding and Joining
• /
• 제16권5호
• /
• pp.119-133
• /
• 1998

To implement full automation in pipe welding, it si most important to develop special sensors and their related systems which act like human operator when detecting irregular groove conditions. In this study, an automatic pipe Gas Metal Arc Welding (GMAW) system was proposed to full control pipe welding procedure with intelligent sensor systems. A five-axes manipulator was proposed for welding torch to automatically access to exact welding position when pipe size and welding angle were given. Pool status and torch position were measured by using a weld-pool image monitoring and processing technique in root-pass welding for weld seam tracking and weld pool control. To overcome the intensive arc light, pool image was captured at the instance of short circuit of welding power loop. Captured image was processed to determine weld pool shape. For weld seam tracking, the relative distance of a torch position from the pool center was calculated in the extracted pool shape to move torch just onto the groove center. To control penetration of root pas, gap was calculated in the extracted pool image, and then weld conditions were controlled for obtaining appropriate penetration. welding speed was determined with a fuzzy logic, and welding current and voltage were determined from a data base to correspond to the gap. For automatic fill-pass welding, the function of human operator of real time weld seam control can be substituted by a sensor system. In this study, an arc sensor system was proposed based on a fuzzy control logic. Using the proposed automatic system, root-pass welding of pipe which had gap variation was assured to be appropriately controlled in welding conditions and in torch position by showing sound welding result and good seam tracking capability. Fill-pass welding by the proposed system also showed very successful result by tracking along the offset welding line without any control of human operator.

• 한국멀티미디어학회논문지
• /
• 제25권2호
• /
• pp.374-381
• /
• 2022

This study analyzes the visual perceptual difference of observers in the artworks created by human artists and artificial intelligence(AI) through eye-tracking. More specifically, the study analyzes the degree of visual attention through a fixation experiment on non-linguistic sources such as the formation and expression of artworks. As a result of this study, the subjects had guessed that one out of four artworks were created by AI (in actuality, 61.1% of the artworks were created by The Next Rembrandt). This demonstrates that most of the subjects hardly recognized the difference between the artwork of human artists and AI. From the comparative analysis of visual perceptual differences found through eye-tracking, more visual attention was found to be demanded for catching details of more stimulating visuals compared to less stimulating visuals. In the gender difference analysis, both of the female and male subjects were likely to stare more intently at the flowers of still-life paintings (Deep Dream & Vincent Van Gogh) while the eyes of a portrait painting (Rembrandt & The Next Rembrandt); this demonstrates no significant differences in gender. Various opinions on AI and art creation from different perspectives arose, therefore, this research is meaningful in a way that it suggests an objective examination through experiments with an artistic perspective.

• Journal of Electrical Engineering and Technology
• /
• 제11권6호
• /
• pp.1857-1862
• /
• 2016

Human activity recognition using depth information is an emerging and challenging technology in computer vision due to its considerable attention by many practical applications such as smart home/office system, personal health care and 3D video games. This paper presents a novel framework of 3D human body detection, tracking and recognition from depth video sequences using spatiotemporal features and modified HMM. To detect human silhouette, raw depth data is examined to extract human silhouette by considering spatial continuity and constraints of human motion information. While, frame differentiation is used to track human movements. Features extraction mechanism consists of spatial depth shape features and temporal joints features are used to improve classification performance. Both of these features are fused together to recognize different activities using the modified hidden Markov model (M-HMM). The proposed approach is evaluated on two challenging depth video datasets. Moreover, our system has significant abilities to handle subject's body parts rotation and body parts missing which provide major contributions in human activity recognition.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.749-752
• /
• 1998

Recognizing a human part becomes very important for applications which are based on the interaction between computers and their users. In this paper, we design and implement a system which recognizes and tracks a human head using a sequence of images. Difference images are used to easily extract feature vectors from images with very complex backgrounds. A human bhead is represented with an ellipse and recognized by searching for a maximum value from preprocessed gradient images. The method is developed by considering the fact that the tracking system should be real-time. The designed system not only shows an excellent performance for the normal up-right position of the head, but also for the cases of 360.deg. rotated head position, occluded images of heads, and tilted head positions.

• ETRI Journal
• /
• 제40권5호
• /
• pp.653-663
• /
• 2018

Patients with spinal cord injuries cannot move their limbs using their intact muscles. A suitable controller can be used to move their arms by employing the functional electrical stimulation method. In this article, a fuzzy exponential sliding-mode controller is designed to move a musculoskeletal human arm model to track an optimal trajectory in the sagittal plane. This optimal arm trajectory is obtained by developing a policy for the central nervous system. In order to specify the optimal trajectory between two points, two dynamic and static optimal criteria are applied simultaneously. The first dynamic objective function is defined to minimize the joint torques, and the second static optimization is offered to minimize the muscle forces at each moment. In addition, fuzzy logic is used to tune the sliding-surface parameter to enable an appropriate tracking performance. Simulation results are evaluated and compared with experimental data for upward and downward movements of the human arm.

• 반도체디스플레이기술학회지
• /
• 제18권1호
• /
• pp.97-101
• /
• 2019

Understanding and classification of the human's emotion play an important tasks in interacting with human and machine communication systems. This paper proposes a novel emotion recognition method by extracting facial keypoints, which is able to understand and classify the human emotion, using active Appearance Model and the proposed classification model of the facial features. The existing appearance model scheme takes an expression of variations, which is calculated by the proposed classification model according to the change of human facial expression. The proposed method classifies four basic emotions (normal, happy, sad and angry). To evaluate the performance of the proposed method, we assess the ratio of success with common datasets, and we achieve the best 93% accuracy, average 82.2% in facial emotion recognition. The results show that the proposed method effectively performed well over the emotion recognition, compared to the existing schemes.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 춘계학술대회
• /
• pp.868-871
• /
• 2009

21세기를 본 논문에서는 칼라분포에 기반한 적응 외형 모델을 파티클 필터에 적용한 이동물체 추적방법을 제시하였다. 칼라 기반의 추적은 서로 다른 외형의 변화에 따라 빠르게 움직이는 이동물체를 다중 관측 모델을 결합하여 추적할 수 방법을 제시하고 있다.

• 한국항행학회논문지
• /
• 제16권3호
• /
• pp.545-552
• /
• 2012

본 논문에서는 사람의 신체 일부분을 추적하는 시스템을 위해서 피부영역을 추출하고 여러 개의 영역을 추적하는 칼만 필터와 가중 탐색 영역을 이용한 다중 CAMShift 알고리즘(KWMCAMShift)을 제안한다. 배경모델을 구성하고 손과 얼굴의 피부색영역을 탐색 영역으로 하는 CAMShift를 제안한다. 이때 CAMShift의 유동적인 탐색영역을 안정화하기 위해 칼만 필터를 이용한다. 손과 얼굴 등이 상호 겹쳐지는 경우 탐색영역의 손실을 막기 위해 주 탐색영역과 비 탐색영역에 대한 가중치를 부가하여 서로 폐색 영역에 대한 회피 알고리즘을 제안한다. 얼굴 영역과 양손의 영역을 중심으로 인간의 자세를 추정하여 어깨와 손과의 관계로 팔꿈치를 추정하였고, 가우시안 배경 모델에 생성되는 그림자를 제거하여 발끝을 찾아 신체 전체를 추정하였다. 제안된 KWMCAMShift 알고리즘을 적용하였을 때 폐색 시에도 96.82%의 인식률을 보였으며 실시간이 가능하였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권1호
• /
• pp.46-51
• /
• 2005

Intelligent Space(ISpace) is the space where many intelligent devices, such as computers and sensors, are distributed. According to the cooperation of many intelligent devices, the environment, it is very important that the system knows the location information to offer the useful services. In order to achieve these goals, we present a method for representing, tracking and human following by fusing distributed multiple vision systems in ISpace, with application to pedestrian tracking in a crowd. And the article presents the integration of color distributions into particle filtering. Particle filters provide a robust tracking framework under ambiguity conditions. We propose to track the moving objects by generating hypotheses not in the image plan but on the top-view reconstruction of the scene. Comparative results on real video sequences show the advantage of our method for multi-object tracking. Simulations are carried out to evaluate the proposed performance. Also, the method is applied to the intelligent environment and its performance is verified by the experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권6호
• /
• pp.2217-2229
• /
• 2015

Since the variations of illumination, the irregular changes of human shapes, and the partial occlusions, multiple person tracking is a challenging work in computer vision. In this paper, we propose a graph clustering method based on spatio-temporal information of moving objects for multiple person tracking. First, the part-based model is utilized to localize individual foreground regions in each frame. Then, we heuristically leverage the spatio-temporal constraints to generate a set of reliable tracklets. Finally, the graph shift method is applied to handle tracklet association problem and consequently generate the completed trajectory for individual object. The extensive comparison experiments demonstrate the superiority of the proposed method.