• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.8
• /
• pp.1553-1559
• /
• 2017

This paper proposes a lane departure detection algorithm using a single camera equipped in front of a vehicle. The proposed algorithm generates a partial top-view image for a small ROI (region of interest) designated on the top-view space form the image acquired by the camera, detects lanes on the small partial top-view image, and makes a decision on the lane departure by checking overlap between the pre-assigned virtual vehicle and the detected lanes. The proposed algorithm also includes the removal of lines occurred by road symbols (noises) disturbing the lane departure detection between lanes and the prediction of lost lanes using lane information of previous fames. In lane departure detection test using real road videos, the proposed algorithm makes the right decision of 99.0% in lane keeping conditions and 94.7% in lane departure conditions.

• Journal of KIISE:Software and Applications
• /
• v.36 no.8
• /
• pp.653-663
• /
• 2009

In this paper, we propose a blind spots elimination system using three cameras. A wide-angle camera is attached on trunk for eliminating blind spots of a rear-view mirror and two cameras are attached on each side-view mirror for eliminating blind spots of vehicle's sides. In order to eliminate blind spots efficiently, we suggest a method to build a panoramic mosaic view with two side images and one wide-angle rear image. First, we obtain an undistorted image from a wide-angle camera of rear-view and calculate the focus-of-contraction (FOC) in undistorted images of rear-view while the car is moving straight forward. Second, we compute a homography among side-view images and an undistorted image of rear-view in flat road scenes. Next, we perform an image registration process after road and background region segmentation. Finally, we generate various views such as a cylinder panorama view, a top view and an information panoramic mosaic view.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.97-106
• /
• 2010

In this paper, lane and curvature detection algorithm based on the curve template matching method is proposed. To eliminate the perspective effect of the original image, the input image is transformed to a top view image. From this top view image, its edge image is created. To increase the accuracy of detection, a novel edge detection method, which shows a strength in lane detection, is proposed. In the first step, straight lanes are detected from the edge image, and then the Curve Template Matching(CTM) method is applied to detect the curved lanes and to find their curvatures. Since the proposed CTM method uses only the simple equations, such as line and circle equations, to detect the curved lane, the algorithm is simple. Moreover, we used the detected lane information in the previous frames to detect the current frame's lanes, the detection results become more reliable. The proposed algorithm has been tested in various road conditions (highway, urban street, night time highway, etc.). Experimental results show that the proposed algorithm can process about 70 frames per second with the successful lane detection rate over 95% and curvature detection rate about 90%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.815-818
• /
• 2016

Forward collision warning systems(FCWS) and lane departure warning systems(LDWS) need regions of interest for detecting lanes and objects as road regions. In general, the lane departure warning system using a vehicle front camera is tracking a lane curve using RANSAC or the like in the form of a straight line obtained image are compared with the center of the vehicle. This algorithm has weaknesses that requires a wide range of the lane being vulnerable to the curve. This paper presents an algorithm that checks whether the current lane departure by car from the Top-view space. The algorithm also can check whether the vehicle in the lane departure of the narrow range, and shows the result that is almost not affected by noise.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.05a
• /
• pp.431-435
• /
• 2006

Panoramic Image provides wider field of view than image from acquisition equipment such as a camera and provides realism and immersion to users compared with single image. Cubic panoramic image provides three dimensional access zooming and rotating in top, bottom, left and right directions. But we require commercial softwares to make a panoramic image and can see distorted images in top and bottom direction. This paper presents a method that constructs cubic panoramic virtual reality image using Apple QuickTimeVR's cubic data structure without any commercial software to make realistic image of top and bottom direction in cubic panoramic virtual reality space.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.11
• /
• pp.1879-1886
• /
• 2016

In this paper, we propose a real-time recognition algorithm of lanes, stop lines and speed bumps on roads for autonomous vehicles. First, we generate a top-view using the image transmitted from a camera that is installed to see the front of a vehicle. To speed up the processing, we simplify the mapping algorithm in constructing a top-view wherein the region of interest (ROI) is concerned. The features of lanes, stop lines and speed bumps, which are composed of lines, are searched in the edge image of the top-view, then followed by labeling and clustering specialized to detect straight lines. The width of lines, distances from the center of a vehicle, and curvature of each cluster are considered to select final candidates. We verify the proposed algorithm on real roads using the commercial car (KIA K7) which is converted into an autonomous vehicle.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.12
• /
• pp.93-102
• /
• 2011

This paper presents a walking guidance system for blind pedestrians in an outdoor environment using just one single camera. Unlike many existing travel-aid systems that rely on stereo-vision, the proposed system aims to get necessary information of the road environment by using just single camera fixed at the belly of the user. To achieve this goal, a top-view image of the road is used, on which obstacles are detected by first extracting local extreme points and then verified by the polar edge histogram. Meanwhile, user motion is estimated by using optical flow in an area close to the user. Based on these information extracted from image domain, an audio message generation scheme is proposed to deliver guidance instructions via synthetic voice to the blind user. Experiments with several sidewalk video-clips show that the proposed walking guidance system is able to provide useful guidance instructions under certain sidewalk environments.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.274-275
• /
• 2008

탑뷰 기반 주자 지원 시스템 (Top-View Based Parking Assistance System)은 차량의 전자장치 기술과 영상 기반 제어 기술을 융합하여 운전자들의 판단 및 조작 실수로 인한 경제적 손실이나 운전자들의 주차에 대한 중압감을 해소한다. 본 논문에서는 이를 구현하기 위하여 차량의 전, 후 좌, 우 네 위치에 카메라를 장착하였고 각 카메라의 높이 초점 거리, 렌즈의 종류와 지면에 대한 각 등을 공식화하여 영강-지면 좌표 모델을 개발하였다. 이를 바탕으로 추출된 영상을 확대, 축소, 회전, 조합 등의 영상처리를 이용하여 차량의 위에서 내려다보는 것과 같은 영상을 생성하는 알고리즘을 제안하고 VC++을 이용하여 소프트웨어 어플리케이션으로 개발하였다.

• Journal of IKEEE
• /
• v.14 no.4
• /
• pp.317-323
• /
• 2010

The video surveillance system is adopted in many places due to its efficiency and constancy in monitoring a specific area over a long period of time. However, many surveillance systems composed of a single static camera often produce unsatisfactory results due to their lack of field of view. In this paper, we present a video surveillance system based on wide baseline stereo cameras to overcome the limitation. We adopt the codebook algorithm and mathematical morphology to robustly model the foreground pixels of the moving object in the scene and calculate the trajectory of the moving object via 3D reconstruction. The experimental results show that the proposed system detects a moving object and generates a top view trajectory successfully to track the location of the object in the world coordinates.

• Journal of Broadcast Engineering
• /
• v.21 no.2
• /
• pp.237-251
• /
• 2016

The conventional 3D-HEVC uses the depth data of the other view instead of that of the current view because the texture data has to be encoded before the corresponding depth data of the current view has been encoded, where the depth data of the other view is used as the predicted depth for the current view. Whereas the conventional 3D-HEVC has no other candidate for the predicted depth information except for that of the other view, the scalable 3D-HEVC utilizes the depth data of the lower spatial layer whose view ID is equal to that of the current picture. The depth data of the lower spatial layer is up-scaled to the resolution of the current picture, and then the enlarged depth data is used as the predicted depth information. Because the quality of the enlarged depth is much higher than that of the depth of the other view, the proposed scheme increases the coding efficiency of the scalable 3D-HEVC codec. Computer simulation results show that the scalable 3D-HEVC is useful and the proposed scheme to use the enlarged depth data for the current picture provides the significant coding gain.