• 한국산업융합학회 논문집
• /
• 제23권2_2호
• /
• pp.255-261
• /
• 2020

We consider the problem of autonomously flying a quadcopter in indoor environments. Navigation in indoor settings poses two major issues. First, real time recognition of the marker captured by the camera. Second, The combination of the distributed images is used to determine the position and orientation of the quadcopter in an indoor environment. We autonomously fly a miniature RC quadcopter in small known environments using an on-board camera as the only sensor. We use an algorithm that combines data-driven image classification with image-combine techniques on the images captured by the camera to achieve real 3D localization and navigation.

• 대한의용생체공학회:의공학회지
• /
• 제27권6호
• /
• pp.392-401
• /
• 2006

Image Guided Surgery (IGS) system which has variously tried in medical engineering fields is able to give a surgeon objective information of operation process like decision making and surgical planning. This information is displayed through 3D images which are acquired from image modalities like CT and MRI for pre-operation. The technique of image registration is necessary to construct IGS system. Image registration means that 3D model and the object operated by a surgeon are matched on the common frame. Major techniques of registration in IGS system have been used by recognizing fiducial markers placed on the object. However, this method has been criticized due to additional trauma, its invasive protocol inserting fiducial markers in patient's bone and generating noise data when 2D slice images are acquired by image modality because many markers are made of metal. Therefore, this paper developed shape-based registration technique to improve the limitation of fiducial marker based IGS system. Iterative Closest Points (ICP) algorithm was used to match corresponding points and quaternion based rotation and translation transformation using closed form solution applied to find the optimized cost function of transformation. we assumed that this algorithm were used in Total Knee replacement (TKR) operation. Accordingly, we have developed region-based 3D registration technique based on anatomical landmarks and this registration algorithm was evaluated in a femur model. It was found that region-based algorithm can improve the accuracy in 3D registration.

• 전자공학회논문지SC
• /
• 제45권3호
• /
• pp.20-31
• /
• 2008

본 논문에서는 단일 카메라를 이용하여 영상 내에 존재하는 객체의 3차원 공간 상에서의 위치 및 높이를 추출하기 위한 기법을 제안한다. 본 논문에서 제안하는 방법은 영상으로 사영된 3차원 장면(scene)에 대한 기준 좌표계를 마커(marker)를 이용해서 설정한 다음, 대상 물체의 2차원 영상을 기준 좌표계로 직접 역사영(back-projection) 시킴으로써 대상 물체에 대한 3차원 공간에서의 위치 및 높이를 계산한다. 그리고 부정확한 카메라 교정으로 인하여 발생하는 역사영 오차를 마커의 기하학 정보를 이용해서 보정한다. 제안된 방법은 기존의 방법에서 주로 이용되던 소실점(vanishing point) 및 소실선(vanishing line) 등을 이용하지 않으며, 3차원 공간 내에서의 객체의 높이 및 위치의 동시 추정이 가능한 장점이 있다. 또한 단일 카메라만을 이용하여 필요한 위치 및 높이 정보를 추출하기 때문에 다중 카메라를 이용한 기법에서 발생할 수 있는 3차원 좌표계 상에서의 대응점의 모호성, 다수의 카메라를 정확히 교정시켜야 하는 어려움 등의 문제가 발생하지 않는다. 실험 결과를 통하여 제안된 기법의 정확도 및 안정성을 확인하였다.

• Radiation Oncology Journal
• /
• 제25권1호
• /
• pp.54-61
• /
• 2007

목 적: 본 연구의 목적은 전립선암 환자의 방사선 치료 시 표적의 정확한 위치를 찾기 위해 표지(marker)를 삽입한 경우 방사선치료계획 시 촬영한 CT 영상과 매 치료 시 온보드 영상장치(on-board imager, OBI)로부터 획득된 직교 kV X선 영상을 이용하여 표지의 위치를 계산하고 자동으로 맞춤을 수행하여 환자 셋업 오차를 보정하도록 하는 방법을 개발하는 것이다. 대상 및 방법: 세 개의 금 표지를 환자 전립선의 기준 위치에 삽입한 후 CT 모의치료기를 이용하여 2 mm 슬라이스 간격으로 CT 영상을 획득하였으며 매 치료 전에 환자 셋업 보정을 위하여 OBI를 이용하여 직교하는 kV X선 환자 영상을 획득하였다. CT 및 kV X선 영상 내 표지 정보 및 좌표 값 추출을 위하여 화소값의 문턱값 처리, 필터링, 외곽선 추출, 패턴 인식 등 다수의 영상처리 알고리듬을 적용하였다. 각 표지들 위치의 대표값으로 삼각형의 무게중심 개념을 이용하였으며 기준 CT 영상 및 직교 kV X선 영상으로부터 각각 무게중심의 좌표를 구한 후 그 차이를 보정해야 할 셋업의 오차로 계산하였다. 알고리듬의 건전성(robustness) 평가를 위해 팬텀을 이용하여 계산된 CT 및 kV X선 영상의 무게중심이 실제 지정된 위치와 일치하는지 여부를 확인하였으며, 본원에서 방사선 치료를 시행한 네 명의 전립선암 환자에 대상으로 치료 직전 촬영한 38 내지 39쌍의 kV X선 영상에 대하여 알고리듬을 적용한 후 OBI 프로그램에서 제공되는 2차원-2차원 맞춤 결과와 비교하였다. 결 과: 팬텀 실험 결과 실제 값과 CT 영상 및 직교 kV X선 영상으로부터 계산된 무게 중심 좌표 값이 1 mm 오차 내에서 일치함을 확인할 수 있었다. 환자 영상에 적용한 경우에도 모든 영상에 대하여 성공적으로 각 표지의 위치를 계산할 수 있었으며 2차원-2차원 맞춤 기능을 이용하여 계산된 셋업 오차와 비교해본 결과 1 mm 범위 내에서 일치함을 확인할 수 있었다. 본 알고리듬을 이용하여 계산한 결과 셋업 오차는 전후(AP) 방향으로 환자별로 작게는 $0.1{\pm}2.7\;mm$에서 크게는 $1.8{\pm}6.6\;mm$까지, 상하(SI) 방향으로 $0.8{\pm}1.6\;mm$에서 $2.0{\pm}2.7\;mm$, 좌우(Lat) 방향으로 $-0.9{\pm}1.5\;mm$에서 $2.8{\pm}3.0\;mm$까지였으며 환자에 따라 그 편차의 차이가 있었다. 결 론: 제안된 알고리듬을 이용하여 1회 셋업 오차를 평가하는 데 소요되는 시간은 10초 미만으로서 임상 적용 시 환자 셋업 시간을 줄이고 주관성을 배제하는 데 도움이 될 수 있을 것으로 기대된다. 그러나 온라인 환자 셋업 보정 시스템에 적용하기 위해서는 선형가속기의 제어 시스템에 통합되는 것이 필요하다.

• 토지주택연구
• /
• 제11권3호
• /
• pp.93-101
• /
• 2020

Recently, as a BIM-based construction simulation system, 4D CAD tools using virtual reality (VR) objects are being applied in construction project. In such a system, since the expression of the object is based on VR image, it has a sense of separation from the real environment, thus limiting the use of field engineers. For this reason, there are increasing cases of applying augmented reality (AR) technology to reduce the sense of separation from the field and express realistic VR objects. This study attempts to develop a methodology and BIM module for the pre-construction verification system using AR technology to increase the practical utility of VR-based BIM objects. To this end, authors develop an AR-based drawing verification function and drawing object-based 4D model augmentation function that can increase the practical utility of 2D drawings, and verify the applicability of the system by performing case analysis. Since VR object-based image has a problem of low realism to field engineers, the linking technology between AR object and 4D model is expected to contribute to the expansion of the use of 4D CADsystem in the construction project.

• 인터넷정보학회논문지
• /
• 제21권1호
• /
• pp.33-43
• /
• 2020

Retrieving a 3D model from a 3D database and augmenting the retrieved model in the Augmented Reality system simultaneously became an issue in developing the plausible AR environments in a convenient fashion. It is considered that the sketch-based 3D object retrieval is an intuitive way for searching 3D objects based on human-drawn sketches as query. In this paper, we propose a novel deep learning based approach of retrieving a sketch-based 3D object as for an Augmented Reality Model. For this work, we introduce a new method which uses Sketch CNN, Wasserstein CNN and Wasserstein center loss for retrieving a sketch-based 3D object. Especially, Wasserstein center loss is used for learning the center of each object category and reducing the Wasserstein distance between center and features of the same category. The proposed 3D object retrieval and augmentation consist of three major steps as follows. Firstly, Wasserstein CNN extracts 2D images taken from various directions of 3D object using CNN, and extracts features of 3D data by computing the Wasserstein barycenters of features of each image. Secondly, the features of the sketch are extracted using a separate Sketch CNN. Finally, we adopt sketch-based object matching method to localize the natural marker of the images to register a 3D virtual object in AR system. Using the detected marker, the retrieved 3D virtual object is augmented in AR system automatically. By the experiments, we prove that the proposed method is efficiency for retrieving and augmenting objects.

• Computers and Concrete
• /
• 제33권5호
• /
• pp.535-544
• /
• 2024

In the rapidly advancing landscape of computer vision (CV) technology, there is a burgeoning interest in its integration with the construction industry. Camera calibration is the process of deriving intrinsic and extrinsic parameters that affect when the coordinates of the 3D real world are projected onto the 2D plane, where the intrinsic parameters are internal factors of the camera, and extrinsic parameters are external factors such as the position and rotation of the camera. Camera pose estimation or extrinsic calibration, which estimates extrinsic parameters, is essential information for CV application at construction since it can be used for indoor navigation of construction robots and field monitoring by restoring depth information. Traditionally, camera pose estimation methods for cameras relied on target objects such as markers or patterns. However, these methods, which are marker- or pattern-based, are often time-consuming due to the requirement of installing a target object for estimation. As a solution to this challenge, this study introduces a novel framework that facilitates camera pose estimation using standardized materials found commonly in construction sites, such as concrete forms. The proposed framework obtains 3D real-world coordinates by referring to construction materials with certain specifications, extracts the 2D coordinates of the corresponding image plane through keypoint detection, and derives the camera's coordinate through the perspective-n-point (PnP) method which derives the extrinsic parameters by matching 3D and 2D coordinate pairs. This framework presents a substantial advancement as it streamlines the extrinsic calibration process, thereby potentially enhancing the efficiency of CV technology application and data collection at construction sites. This approach holds promise for expediting and optimizing various construction-related tasks by automating and simplifying the calibration procedure.

• ETRI Journal
• /
• 제33권2호
• /
• pp.259-266
• /
• 2011

We present a new method for an automated markerless system to describe, analyze, and classify human gait motion. The automated system consists of three stages: I) detection and extraction of the moving human body and its contour from image sequences, ii) extraction of gait figures by the joint angles and body points, and iii) analysis of motion parameters and feature extraction for classifying human gait. A sequential set of 2D stick figures is used to represent the human gait motion, and the features based on motion parameters are determined from the sequence of extracted gait figures. Then, a k-nearest neighbor classifier is used to classify the gait patterns. In experiments, this provides an alternative estimate of biomechanical parameters on a large population of subjects, suggesting that the estimate of variance by marker-based techniques appeared generous. This is a very effective and well-defined representation method for analyzing the gait motion. As such, the markerless approach confirms uniqueness of the gait as earlier studies and encourages further development along these lines.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제11권2호
• /
• pp.679-687
• /
• 2019

This paper presents the autonomous swimming technology developed for an Autonomous Underwater Vehicle (AUV) operating in the underwater jacket structure environment. To prevent the position divergence of the inertial navigation system constructed for the primary navigation solution for the vehicle, we've developed kinds of marker-recognition based underwater localization methods using both of optical and acoustic cameras. However, these two methods all require the artificial markers to be located near to the cameras mounted on the vehicle. Therefore, in the case of the vehicle far away from the structure where the markers are usually mounted on, we may need alternative position-aiding solution to guarantee the navigation accuracy. For this purpose, we develop a sonar image processing based underwater localization method using a Forward Looking Sonar (FLS) mounted in front of the vehicle. The primary purpose of this FLS is to detect the obstacles in front of the vehicle. According to the detected obstacle(s), we apply an Occupancy Grid Map (OGM) based path planning algorithm to derive an obstacle collision-free reference path. Experimental studies are carried out in the water tank and also in the Pohang Yeongilman port sea environment to demonstrate the effectiveness of the proposed autonomous swimming technology.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제10권2호
• /
• pp.67-80
• /
• 2006

본 논문에서는 스테레오 비전을 이용하여 기존의 포인팅 디바이스인 마우스를 대신할 새로운 포인팅 디바이스 방법을 제안 한다. 제안한 방법은 기존의 인식 장비들이 가진 마커에 의한 움직임 제약과 고가 장비의 단점을 극복하기 위해 컴퓨터 비전을 이용한다. 즉, 사람이 시각을 통해 정보를 인식하는 것과 동일하게 컴퓨터의 영상 정보를 이용하여 실시간으로 영상내의 컬러 영역의 분할을 통해 물체의 위치를 추적 및 정합하고 이의 위치를 스테레오 기하학 정보로부터 계산하여 포인팅 동작을 수행한다.