• Corrosion Science and Technology
• /
• v.18 no.6
• /
• pp.253-257
• /
• 2019

A deep learning based real-time painting surface inspection algorithm is proposed herein, designed for developing an autonomous inspection drone. The painting surface inspection is usually conducted manually. However, the manual inspection has a limitation in obtaining accurate data for correct judgement on the surface because of human error and deviation of individual inspection experiences. The best method to replace manual surface inspection is the vision-based inspection method with a camera, using various image processing algorithms. Nevertheless, the visual inspection is difficult to apply to surface inspection due to diverse appearances of material, hue, and lightning effects. To overcome technical limitations, a deep learning-based pattern recognition algorithm is proposed, which is specialized for painting surface inspections. The proposed algorithm functions in real time on the embedded board mounted on an autonomous inspection drone. The inspection results data are stored in the database and used for training the deep learning algorithm to improve performance. The various experiments for pre-inspection of painting processes are performed to verify real-time performance of the proposed deep learning algorithm.

• Proceedings of the IEEK Conference
• /
• 2004.08c
• /
• pp.862-866
• /
• 2004

Non-intrusive methods based on active remote IR illumination for eye tracking is important for many applications of vision-based man-machine interaction. One problem that has plagued those methods is their sensitivity to lighting condition change. This tends to significantly limit their scope of application. In this paper, we present a new real-time eye detection and tracking methodology that works under variable and realistic lighting conditions. Based on combining the bright-pupil effect resulted from IR light and the conventional appearance-based object recognition technique, our method can robustly track eyes when the pupils are not very bright due to significant external illumination interferences. The appearance model is incorporated in both eyes detection and tracking via the use of support vector machine and the mean shift tracking. Additional improvement is achieved from modifying the image acquisition apparatus including the illuminator and the camera.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.2
• /
• pp.240-244
• /
• 2007

Non-intrusive methods based on active remote IR illumination fur eye tracking is important for many applications of vision-based man-machine interaction. One problem that has plagued those methods is their sensitivity to lighting condition change. This tends to significantly limit their scope of application. In this paper, we present a new real-time eye detection and tracking methodology that works under variable and realistic lighting conditions. Based on combining the bright-pupil effect resulted from IR light and the conventional appearance-based object recognition technique, our method can robustly track eyes when the pupils are not very bright due to significant external illumination interferences. The appearance model is incorporated in both eyes detection and tracking via the use of support vector machine and the mean shift tracking. Additional improvement is achieved from modifying the image acquisition apparatus including the illuminator and the camera.

• The Journal of Korea Robotics Society
• /
• v.11 no.4
• /
• pp.205-216
• /
• 2016

The localization of the robot is one of the most important factors of navigating mobile robots. The use of featured information of landmarks is one approach to estimate the location of the robot. This approach can be classified into two categories: the natural-landmark-based and artificial-landmark-based approach. Natural landmarks are suitable for any environment, but they may not be sufficient for localization in the less featured or dynamic environment. On the other hand, artificial landmarks may generate shaded areas due to space constraints. In order to improve these disadvantages, this paper presents a novel development of the localization system by using artificial and natural-landmarks-based approach on a topological map. The proposed localization system can recognize far or near landmarks without any distortion by using landmark tracking system based on top-view image transform. The camera is rotated by distance of landmark. The experiment shows a result of performing position recognition without shading section by applying the proposed system with a small number of artificial landmarks in the mobile robot.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.2
• /
• pp.456-463
• /
• 2004

Non-intrusive methods based on active remote IR illumination for eye tracking is important for many applications of vision-based man-machine interaction. One problem that has plagued those methods is their sensitivity to lighting condition change. This tends to significantly limit their scope of application. In this paper, we present a new real-time eye detection and tacking methodology that works under variable and realistic lighting conditions. Based on combining the bright-Pupil effect resulted from IR light and the conventional appearance-based object recognition technique, our method can robustly track eyes when the pupils ale not very bright due to significant external illumination interferences. The appearance model is incorporated in both eyes detection and tacking via the use of support vector machine and the mean shift tracking. Additional improvement is achieved from modifying the image acquisition apparatus including the illuminator and the camera.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.300-303
• /
• 2007

지문자는 청각장애인이 사용하는 수화로 표현하지 못하는 한글 문자를 알파벳으로 표시하기위한 손 제스처이다. 본 논문에서는 추출된 손 영역의 무게 중심과 퍼지 논리를 이용하여 지문자를 인식하는 알고리즘을 제안하고, 한글 문자를 표현하는 시스템을 개발한다. USB 카메라로부터 얻어진 영상에서 히스토그램을 이용하여 손의 피부색 영역을 추출하고, 영상 마스크를 이용하여 피부색이 아닌 배경 영역을 제거한다. 문턱 값을 사용하여 얻어진 이진화된 영상에서 손의 영역을 검출하고, 무게 중심을 이용하여 손 중심과 손가락 끝의 거리를 측정한다. 얻어진 거리 정보에 퍼지 기법을 적용하여 손가락의 굽힘 정도를 판단하고, 손 모양 데이터베이스에서 손가락 굽힘 정도와 가장 근사한 한글 문자를 선택한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.652-655
• /
• 2012

With the development of IT and digital camera technology, diverse applications of image processing services are becoming available. In this paper, in order to ultimately automate parking management system, we designed and implemented a system for recognizing vehicle license plates that show vehicles' unique numbers by using EmguCV which shows optimized performance on Intel-based environment. We also implemented UI for administrators to easily manage the entire system by utilizing.

• ETRI Journal
• /
• v.33 no.1
• /
• pp.78-88
• /
• 2011

In this paper, we present the Touch Text exTractor (Touch TT), an interactive text segmentation tool for the extraction of scene text from camera-based images. Touch TT provides a natural interface for a user to simply indicate the location of text regions with a simple touchline. Touch TT then automatically estimates the text color and roughly locates the text regions. By inferring text characteristics from the estimated text color and text region, Touch TT can extract text components. Touch TT can also handle partially drawn lines which cover only a small section of text area. The proposed system achieves reasonable accuracy for text extraction from moderately difficult examples from the ICDAR 2003 database and our own database.

• International Journal of Contents
• /
• v.9 no.4
• /
• pp.16-21
• /
• 2013

Vision system is an indispensable part of constructing the chess-playing robot. Chessboard detection and pieces localization in the captured image of robot's camera are important steps for processes followed such as pieces recognition, move calculation, and robot controlling. We present a method for detecting the Janggi chessboard and pieces based on the edge and color feature. Hough transform combined with line extraction is used for segmenting the chessboard and warping it to form the rectangle shape in order to detect and interpolate the lines of chessboard. Then we detect the existence of pieces and their side by applying the saliency map and checking the color distribution at piece locations. While other methods either work only with the empty chessboard or do not care about the piece existence, our method could detect sufficiently side and position of pieces as well as lines of the chessboard even if the occlusion happens.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2012.05a
• /
• pp.22-25
• /
• 2012

본 논문에서는 키보드나 마우스와 같은 인풋 디바이스 없이 손동작만으로 컴퓨터를 조작할 수 있는 차세대 인터페이스인 에어 인터페이스를 구현하였다. 이 시스템 구현을 위해 먼저 적외선 카메라와 적외선 LED를 사용하였고, 또한 적외선의 전반사의 원리를 이용하였다. 이를 통해 획득된 적외선 영상은 다양한 영상처리 기술을 통해 손 영역을 분할, 검출하게 되고, 검출된 정보를 이용하여 다양한 메타포를 구현 하였으며, 최종적으로 개별적인 제어 이벤트에 각각 맵핑되어 컴퓨터를 제어 및 동작하는 과정을 수행하게 된다. 본 연구에서는 손 영역의 검출과 추적, 인식과정 등을 자세히 서술하였으며, 다양한 메타포도 함께 소개가 된다. 그리고 본 연구에서 사용된 기술을 더욱 발전시키면 사회 전반적으로 다양한 부분에서 적극 활용될 것으로 기대된다.