• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1271-1281
• /
• 2017

It is essential to collect and analyze target information around the vehicle for autonomous driving of the vehicle. Based on the analysis, environmental information such as location and direction should be analyzed in real time to control the vehicle. In particular, obstruction or cutting of objects in the image must be handled to provide accurate information about the vehicle environment and to facilitate safe operation. In this paper, we propose a method to simultaneously generate 2D and 3D bounding box proposals using LiDAR Edge generated by filtering LiDAR sensor information. We classify the classes of each proposal by connecting them with Region-based Fully-Covolutional Networks (R-FCN), which is an object classifier based on Deep Learning, which uses two-dimensional images as inputs. Each 3D box is rearranged by using the class label and the subcategory information of each class to finally complete the 3D bounding box corresponding to the object. Because 3D bounding boxes are created in 3D space, object information such as space coordinates and object size can be obtained at once, and 2D bounding boxes associated with 3D boxes do not have problems such as occlusion.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.510-513
• /
• 2016

Detecting of the object in image processing is substantial but it depends on the object itself and the environment. An object can be detected either by its shape or color. Color is an essential for pattern recognition and computer vision. It is an attractive feature because of its simplicity and its robustness to scale changes and to detect the positions of the object. Generally, color of an object depends on its characteristics of the perceiving eye and brain. Physically, objects can be said to have color because of the light leaving their surfaces. Here, we conducted experiment in the aquarium fish tank. Different color of fish robots are mimic the natural swim of fish. Unfortunately, in the underwater medium, the colors are modified by attenuation and difficult to identify the color for moving objects. We consider the fish motion as a moving object and coordinates are found at every instinct of the aquarium to detect the position of the fish robot using OpenCV color detection. In this paper, we proposed to identify the position of the fish robot by their color and use the position data to control the fish robot gathering in one point in the fish tank through serial communication using RF module. It was verified by the performance test of detecting the position of the fish robot.

• 한국정밀공학회지
• /
• 제17권8호
• /
• pp.60-70
• /
• 2000

In this paper a 3D position measuring device that finds the 3D position of an arbitarily placed object using a camersa system is introduced. The camera system consists of three stepping motors and a CCD camera and a laser. The viewing direction of the camera is controlled by two stepping motors (pan and tilt motors) and the direction of a laser is also controlled by a stepping motors(laser motor). If an object in a remote place is selected from a live video image the x,y,z coordinates of the object with respect to the reference coordinate system can be obtained by calculating the distance from the camera to the object using a structured light scheme and by obtaining the orientation of the camera that is controlled by two stepping motors. The angles o f stepping motors are controlled by a SGI O2 workstation through a parallel port. The mathematical model of the camera and the distance measuring system are calibrated to calculate an accurate position of the object. This 3D position measuring device can be used to acquire information that is necessary to monitor a remote place.

• 한국CDE학회논문집
• /
• 제9권4호
• /
• pp.397-405
• /
• 2004

Texture mapping is the process of covering 3D models with texture images in order to increase the visual realism of the models. For proper mapping the coordinates of texture images need to coincide with those of the 3D models. When projective images from the camera are used as texture images, the texture image coordinates are defined by a camera calibration method. The texture image coordinates are determined by the relation between the coordinate systems of the camera image and the 3D object. With the projective camera images, the distortion effect caused by the camera lenses should be compensated in order to get accurate texture coordinates. The distortion effect problem has been dealt with iterative methods, where the camera calibration coefficients are computed first without considering the distortion effect and then modified properly. The methods not only cause to change the position of the camera perspective line in the image plane, but also require more control points. In this paper, a new iterative method is suggested for reducing the error by fixing the principal points in the image plane. The method considers the image distortion effect independently and fixes the values of correction coefficients, with which the distortion coefficients can be computed with fewer control points. It is shown that the camera distortion effects are compensated with fewer numbers of control points than the previous methods and the projective texture mapping results in more realistic image.

• 한국농촌건축학회논문집
• /
• 제15권2호
• /
• pp.9-16
• /
• 2013

This paper is to discover how the photogrammetry method can be utilized for recording modern architectural properties. As a result of conducting photogrammetry aimed at the main building and water supply tower of Goesan Middle School, it was possible to measure actually either complicated patterns or a damaged area on the facade. However, there appeared errors in the actual measurement of the hollow or protruded area on the facade to measure actually or of an object in the cylinder form. First, for the actual measurement of the protruded area, it is possible to complete a facade by measuring actually the protruded area separately after dividing it and complementing it on the facade, Second, concerning an object in the cylinder form, it is possible to complete one facade by matching a facade which is measured actually with a Diaver method among the MSR photogrammetry methods, and three dimensional standard coordinates which are obtained through Totalstation.

• 한국시뮬레이션학회:학술대회논문집
• /
• 한국시뮬레이션학회 2001년도 The Seoul International Simulation Conference
• /
• pp.248-254
• /
• 2001

This paper presents a progressive algorithm for reconstructing a 3D structure from a given 2D sketch drawing (edge-vertex graph without hidden line removal) according to the user's sketch order. While previous methods reconstruct a 3D structure at once, the proposed method progressively calculate a 3D structure by optimizing the coordinates of vertices of an object according to the sketch order. The progressive method reconstructs the most plausible 3D object quickly by applying 3D constraints that are derived from the relationship between the object and the sketch drawing in the optimization process. The progressive reconstruction algorithm is discussed, and examples from a working implementation are given.

• 한국정밀공학회지
• /
• 제12권12호
• /
• pp.38-44
• /
• 1995

This paper presents a method to measure the position and orientation of a Cylindrical Object(unknown the eiameter and length) lying on a floor, using a camera. The two extreme cross section of the cylinder will be viewed as distorted ellipese or circular are, while its limb edge will be shown as two straight lines. The diameter of the cylinder is determined from the geometric properties of the two straight lines, which in turn provides information regarding the length of the cylinder. From the 3-dimensional measurement, the 3D coordinates of the center points of the two extreme cross sections are determined to give the position and orientation of the cylinder. This method is used for automated pick-and-place operations of cylinder, such as sheet coils, or drums in warehouses.

• 정보처리학회논문지B
• /
• 제17B권6호
• /
• pp.429-436
• /
• 2010

본 논문에서는 단일 카메라로부터 획득한 영상으로부터 물체의 3차원 윤곽선을 구성하는 방법을 제안한다. MOPS(Multi-Scale Oriented Patches) 알고리즘을 이용하여 물체의 대략적인 윤곽선을 검출하고 윤곽선 위에 분포한 특징점의 공간좌표를 획득한다. 동시에 SIFT(Scale Invariant Feature Transform) 알고리즘을 통하여 물체의 윤곽선 내부에 존재하는 특징점 공간좌표를 획득한다. 이러한 정보를 병합하여 물체의 전체 3차원 윤곽선 정보를 구성한다. 본 논문에서 제안하는 방법은 대략적인 물체의 윤곽선만 구성하기 때문에 빠른 계산이 가능하며 SIFT 특징점을 통해 윤곽선 내부 정보를 보완하기 때문에 물체의 자세한 3차원 정보를 얻을 수 있는 장점이 있다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2003년도 춘계학술발표논문집 (상)
• /
• pp.409-412
• /
• 2003

The 3-D shape use to effect of movie, animation, industrial design, medical treatment service, education, engineering etc... But it is not easy to make 3-D shape from the information of 2-D image. There are two methods in restoring 3-D video image through 2-D image; First the method of using a laser; Second, the method of acquiring 3-D image through stereo vision. Instead of doing two methods with many difficulties, I study the method of simple 3-D image in this research paper. We present here a simple and efficient method, called direct calibration, which does not require any equations at all. The direct calibration procedure builds a lookup table(LUT) linking image and 3-D coordinates by a real 3-D triangulation system. The LUT is built by measuring the image coordinates of a grid of known 3-D points, and recording both image and world coordinates for each point; the depth values of all other visible points are obtained by interpolation.

• 한국정보기술학회논문지
• /
• 제17권8호
• /
• pp.75-83
• /
• 2019

본 논문에서는 Median flow와 영상의 Depth 정보를 이용하여 산업 현장의 다양한 환경 조건에서 실시간 철근의 끝점 추적 및 검출이 가능한 방법을 제안한다. 영상의 Depth 정보에 Median filter, Binarization, Morphology, Blob의 알고리즘을 사용하여 2개의 철근 끝점을 검출하는 방법을 제안하였다. 실시간 철근 끝점 추적을 위해서는 Median flow의 알고리즘을 이용하여 철근의 움직임 방향과 위치 추적을 제안하였다. 그리고 Depth 영상에서 검출된 철근 끝점 좌표와 Median flow를 이용한 철근의 위치추적 좌표를 서로 비교하여 최종 위치 좌표를 결정하였다. 그 결과 기존 Median flow 방식만 적용하였을 때 75% 정도의 철근의 최종 위치 판단 성공률이 Depth의 철근 끝점 정보까지 활용하였을 때는 95%까지 추적 성공률이 높아졌다.