• Journal of Welding and Joining
• /
• 제22권6호
• /
• pp.43-49
• /
• 2004

In this study, a visual sensor system which can detect I-butt weld joint with height variation and includes a seam tracking algorithm was investigated. Three-dimensional position of an object can be acquired by using the method of distance measurement, i.e., an optical trigonometry which results from the spatial relations between the camera, the object and the structured light by a visible laser. Effects of laser intensity and iris number for the image quality as well as object material were investigated for the optical system design. For the image processing, a region of interest is defined from the whole image and a line image of laser is drew by using the gray level difference in the image. From the drew laser line, the weld joint can be recognized in searching the biggest point position calculated from the central difference method. Through a series of welding experiments, a good tracking performance was confirmed under GMA welding.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.4929-4947
• /
• 2017

In visual object tracking, Correlation Filter-based Tracking (CFT) systems have arouse recently to be the most accurate and efficient methods. The CFT's circularly shifts the larger search window to find most likely position of the target. The need of larger search window to cover both background and object make an algorithm sensitive to the background and the target occlusions. Further, the use of fixed-sized windows for training makes them incapable to handle scale variations during tracking. To address these problems, we propose two layer target representation in which both global and local appearances of the target is considered. Multiple local patches in the local layer provide robustness to the background changes and the target occlusion. The target representation is enhanced by employing additional reversed RGB channels to prevent the loss of black objects in background during tracking. The final target position is obtained by the adaptive weighted average of confidence maps from global and local layers. Furthermore, the target scale variation in tracking is handled by the statistical model, which is governed by adaptive constraints to ensure reliability and accuracy in scale estimation. The proposed structural enhancement is tested on VTBv1.0 benchmark for its accuracy and robustness.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.72-81
• /
• 2022

Visual Object Tracking is known as the most fundamental problem in the field of computer vision. Object tracking localize the region of target object with bounding box in the video. In this paper, a custom CNN is created to extract object feature that has strong and various information. This network was constructed as a Siamese network for use as a feature extractor. The input images are passed convolution block composed of a bottleneck layers, and features are emphasized. The feature map of the target object and the search area, extracted from the Siamese network, was input as a local proposal network. Estimate the object area using the feature map. The performance of the tracking algorithm was evaluated using the OTB2013 dataset. Success Plot and Precision Plot were used as evaluation matrix. As a result of the experiment, 0.611 in Success Plot and 0.831 in Precision Plot were achieved.

• 제어로봇시스템학회논문지
• /
• 제17권9호
• /
• pp.902-907
• /
• 2011

In this paper, we deal with performance evaluation of visual path following using 2D and 3D information. Visual path follow first teaches driving path by selecting milestone images then follows the same route by comparing the milestone image and current image. We follow the visual path following algorithm of [8] and [10]. In [8], a robot navigated with 2D image information only. But in [10], local 3D geometries are reconstructed between the milestone images in order to achieve fast feature prediction which allows the recovery from tracking failures. Experimental results including diverse indoor cases show performance of each algorithm.

• 로봇학회논문지
• /
• 제18권3호
• /
• pp.260-270
• /
• 2023

We introduce a novel Visual Inertial Odometry (VIO) algorithm designed to improve the performance of thermal-inertial odometry. Thermal infrared image, though advantageous for feature extraction in low-light conditions, typically suffers from a high noise level and significant information loss during the 8-bit conversion. Our algorithm overcomes these limitations by approximating a 14-bit raw pixel histogram into a Gaussian mixture model. The conversion method effectively emphasizes image regions where texture for visual tracking is abundant while reduces unnecessary background information. We incorporate the robust learning-based feature extraction and matching methods, SuperPoint and SuperGlue, and zero velocity detection module to further reduce the uncertainty of visual odometry. Tested across various datasets, the proposed algorithm shows improved performance compared to other state-of-the-art VIO algorithms, paving the way for robust thermal-inertial odometry.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권10호
• /
• pp.5112-5128
• /
• 2016

Tracking human body is an important problem in computer vision field. Tracking failures caused by occlusion can lead to wrong rectification of the target position. In this paper, a robust human tracking algorithm is proposed to address the problem of occlusion, rotation and improve the tracking accuracy. It is based on Tracking-Learning-Detection framework. The key auxiliary information is used in the framework which motivated by the fact that a tracking target is usually embedded in the context that provides useful information. First, face localization method is utilized to find key face location information. Second, the relative position relationship is established between the auxiliary information and the target location. With the relevant model, the key face information will get the current target position when a target has disappeared. Thus, the target can be stably tracked even when it is partially or fully occluded. Experiments are conducted in various challenging videos. In conjunction with online update, the results demonstrate that the proposed method outperforms the traditional TLD algorithm, and it has a relatively better tracking performance than other state-of-the-art methods.

• 한국지능시스템학회논문지
• /
• 제14권5호
• /
• pp.539-544
• /
• 2004

본 논문에서는 CONDENSATION 알고리즘을 이용하여 입자 필터(particle filter)에 기반 한 물체 추적 알고리즘을 제안한다. 입자 필터는 조건 확률 전파 모델(Conditional Density Propagation)인 베이지안(Bayesian) 추론 규칙을 적용하는 추적구조를 갖고 있기 때문에 다른 어떤 종류의 추적 알고리즘보다 뛰어난 성능을 보인다. 논문에서는 실험 결과를 통해, 외곽(contour) 추적 입자 필터가 복잡한 환경 속에서 강인한 추적 성능을 나타냄을 증명한다.

• 방송공학회논문지
• /
• 제21권6호
• /
• pp.913-919
• /
• 2016

본 논문에서는 다중 랜덤 워커(multiple random walkers)에 기반한 객체 추적 기법을 제안한다. 우선 서포트 벡터 머신(support vector machine)을 이용한 분류기 기반 객체 추적 기법을 소개한다. 다음으로 영상의 영역에 대한 특징 벡터 중 배경으로부터 추출된 특징 벡터를 억제하는 기법을 제안한다. 영역에서 배경 요소를 찾기 위해 다중 랜덤 워커를 이용한 전경 및 배경 추출 방법을 제시한다. 배경 요소가 억제된 특징 벡터를 이용하여 학습된 서포트 벡터 머신은 객체와 배경이 유사한 영상, 객체가 다른 물체에 의해 가려지는 영상 등에서 객체와 배경을 확실하게 구분하고, 추적 알고리즘은 정확한 객체 추적을 수행한다. 또한, 객체 추적 알고리즘의 응용에서 중요한 속도 문제를 크게 개선하는 방법을 제안한다. 마지막으로 실험을 통해 제안하는 기법이 높은 처리 속도를 유지하면서 동시에 기존 기법보다 우수한 추적 성능을 보임을 확인한다.

• 한국항공운항학회지
• /
• 제25권4호
• /
• pp.83-90
• /
• 2017

Unmanned Aerial Vehicles(UAVs) require collision avoidance capabilities equivalent to the capabilities of manned aircraft to enter the airspace of manned aircraft. In the case of Visual Flight Rules of manned aircraft, collision avoidance is performed by 'See-and-Avoid' of pilots. To obtain those capabilities of UAVs named as 'Sense-and-Avoid', sensor-system-based intruder tracking and collision avoidance methods are required. In this study, a multi-sensor-based tracking, data fusion, and collision avoidance algorithm is designed by using a model-based design tool MATLAB/SIMULINK, and validations of the designed model and code using numerical simulations and processor-in-the-loop simulations are performed.

• 동력기계공학회지
• /
• 제6권1호
• /
• pp.96-101
• /
• 2002

In this paper, we address the problem of controlling an end-effector to track and grab a moving target using the visual servoing technique. A visual servo mechanism based on the image-based servoing principle, is proposed by using visual feedback to control an end-effector without calibrated robot and camera models. Firstly, we consider the control problem as a nonlinear least squares optimization and update the joint angles through the Taylor Series Expansion. And to track a moving target in real time, the Jacobian estimation scheme(Dynamic Broyden's Method) is used to estimate the combined robot and image Jacobian. Using this algorithm, we can drive the objective function value to a neighborhood of zero. To show the effectiveness of the proposed algorithm, simulation results for a six degree of freedom robot are presented.