• 로봇학회논문지
• /
• 제9권3호
• /
• pp.170-177
• /
• 2014

This paper proposes a novel upward-looking camera-based global localization using a ceiling image map. The ceiling images obtained through the SLAM process are integrated into the ceiling image map using a particle filter. Global localization is performed by matching the ceiling image map with the current ceiling image using SURF keypoint correspondences. The robot pose is then estimated by the coordinate transformation from the ceiling image map to the global coordinate system. A series of experiments show that the proposed method is robust in real environments.

• 로봇학회논문지
• /
• 제3권2호
• /
• pp.99-105
• /
• 2008

To achieve autonomous mobile robot navigation, accurate localization technique is the fundamental issue that should be addressed. In augmented reality, the position of a user is required for location-based services. This paper presents indoor localization using infrared reflective artificial landmarks. In order to minimize the disturbance to the user and to provide the ease of installation, the passive landmarks are used. The landmarks are made of coated film which reflects the infrared light efficiently. Infrared light is not visible, but the camera can capture the reflected infrared light. Once the artificial landmark is identified, the camera's relative position/orientation is estimated with respect to the landmark. In order to reduce the number of the required artificial landmarks for a given environment, the pan/tilt mechanism is developed together with the distortion correction algorithm.

• 한국군사과학기술학회지
• /
• 제12권6호
• /
• pp.768-775
• /
• 2009

In this paper, the sound source localization system in real time which uses the time delay of arrival signal is proposed. This system uses minimum microphones and surveillance camera for estimation of the sound source localization and sound direction. To apply this system to the military field, four models(model1~model4) are derived. Model 1 can be used to evaluate the sound source localization at the long distance. Model2 and model3 can be applied to estimate the sound direction. Model4 is useful for the special purpose and potable device. It is possible for this system to be used for the military guard and surveillance. As a result of experiments, It is shown that this system can estimate the sound source localization and the sound direction using minimum microphones.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.128-131
• /
• 2004

In this paper, we describe reconstructed 3D-urban modeling techniques for laser scanner and CCD camera system, which are loading on the vehicle. We use two laser scanners, the one is horizon scanner and the other is vertical scanner. Horizon scanner acquires the horizon data of building for localization. Vertical scan data are main information for constructing a building. We compared extraction of edge aerial image with laser scan data. This method is able to correct the cumulative error of self-localization. Then we remove obstacles of 3D-reconstructed building. Real-texture information that is acquired with CCD camera is mapped by 3D-depth information. 3D building of urban is reconstructed to 3D-virtual world. These techniques apply to city plan. 3D-environment game. movie background. unmanned-patrol etc.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.404-404
• /
• 2000

This paper presents a single vision-based sel(-localization method in an corridor environment. We use the Hough transform for finding parallel lines and vertical lines. And we use these cross points as feature points and it is calculated relative distance from mobile robot to these points. For matching environment map to feature points, searching window is defined and self-localization is performed by matching procedure. The result shows the suitability of this method by experiment.

• 한국멀티미디어학회논문지
• /
• 제20권5호
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.353-364
• /
• 2009

In this paper, a SLAM (Simultaneous Localization and Mapping) method using a single camera and planar fiducial markers is proposed. Fiducial markers are planar patterns that are mounted on the ceiling or wall. Each fiducial marker has a unique hi-tonal identification pattern with square outlines. It can be printed on paper to reduce cost or it can be painted using retro-reflective paint in order to make invisible and prevent undesirable visual effects. Existing localization methods using artificial landmarks have the disadvantage that landmark locations must be known a priori. In contrast, the proposed method can build a map and estimate robot location even if landmark locations are not known a priori. Hence, it reduces installation time and setup cost. The proposed method works good even when only one fiducial marker is seen at a scene. We perform computer simulation to evaluate proposed method.

• 제어로봇시스템학회논문지
• /
• 제11권8호
• /
• pp.678-687
• /
• 2005

This paper describes a relative localization algorithm using odometry data and consecutive local maps. The purpose of this paper is the odometry error correction using the area matching of two consecutive local maps. The local map is built up using a sensor module with dual laser beams and USB camera. The range data form the sensor module is measured using the structured lighting technique (active stereo method). The advantage in using the sensor module is to be able to get a local map at once within the camera view angle. With this advantage, we propose the AVS (Aligned View Sector) matching algorithm for. correction of the pose error (translational and rotational error). In order to evaluate the proposed algorithm, experiments are performed in real environment.

• 제어로봇시스템학회논문지
• /
• 제18권8호
• /
• pp.744-749
• /
• 2012

In this paper, we suggest that how to improve an accuracy of mobile robot's localization by using the sensor network information which fuses the machine vision camera, encoder and IMU sensor. The heading value of IMU sensor is measured using terrestrial magnetism sensor which is based on magnetic field. However, this sensor is constantly affected by its surrounding environment. So, we isolated template of ceiling using vision camera to increase the sensor's accuracy when we use IMU sensor; we measured the angles by pattern matching algorithm; and to calibrate IMU sensor, we compared the obtained values with IMU sensor values and the offset value. The values that were used to obtain information on the robot's position which were of Encoder, IMU sensor, angle sensor of vision camera are transferred to the Host PC by wireless network. Then, the Host PC estimates the location of robot using all these values. As a result, we were able to get more accurate information on estimated positions than when using IMU sensor calibration solely.

• 제어로봇시스템학회논문지
• /
• 제15권5호
• /
• pp.519-524
• /
• 2009

This paper describes a new sensor system for 3D environment perception using stereo structured infrared light sources and a camera. Environment and obstacle sensing is the key issue for mobile robot localization and navigation. Laser scanners and infrared scanners cover $180^{\circ}$ and are accurate but too expensive. Those sensors use rotating light beams so that the range measurements are constrained on a plane. 3D measurements are much more useful in many ways for obstacle detection, map building and localization. Stereo vision is very common way of getting the depth information of 3D environment. However, it requires that the correspondence should be clearly identified and it also heavily depends on the light condition of the environment. Instead of using stereo camera, monocular camera and two projected infrared light sources are used in order to reduce the effects of the ambient light while getting 3D depth map. Modeling of the projected light pattern enabled precise estimation of the range. Two successive captures of the image with left and right infrared light projection provide several benefits, which include wider area of depth measurement, higher spatial resolution and the visibility perception.