• Journal of Advanced Information Technology and Convergence
• /
• v.10 no.1
• /
• pp.15-23
• /
• 2020

In this paper, we propose a new method to detect traffic accidents in video from vehicle-mounted cameras (vehicle black box). We use the distance between vehicles to determine whether an accident has occurred. To calculate the position of each vehicle, we use object detection and tracking method. By the way, in a crowded road environment, it is so difficult to decide an accident has occurred because of parked vehicles at the edge of the road. It is not easy to discriminate against accidents from non-accidents because a moving vehicle and a stopped vehicle are mixed on a regular downtown road. In this paper, we try to increase the accuracy of the vehicle accident detection by using not only the motion of the surrounding vehicle but also ego-motion as the input of the Recurrent Neural Network (RNN). We improved the accuracy of accident detection compared to the previous method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.773-778
• /
• 1991

In this paper we describe a road following method for an autonomous vehicle. From a road image in gray level, a road boundary is detected using a gradient operator, and then the road boundary is converted to orthogonal view of the road showing the vehicle position and heading direction. In this research an efficient road boundary search technique is developed to support real time vehicle control. Also, an obstacle detection method, using images taken from two different positions, has been developed.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2962-2964
• /
• 1999

We know the traffic information about the velocity and position of vehicle by extraction and tracking vehicle from continuosly obtained road image of camera. The conventional method of vehicle detection indicate increment of error due to headlight and taillight in night road image. This paper show such as vehicle detection of binary, Edge detection. amalgamation of image are applied to extract the vehicle, and Kalman filter is adaptive methods for tracking position and velocity of vehicle.

• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.297-299
• /
• 2006

Searching the region of interest on the Hough transform domain is done to determine the real road boundary on the high speed way. The mathematical morphology is employed to obtain the gradient image which is utilized in Hough transform. Many possible candidates of lines could appear on the ordinary road environment and simple selection of the strongest line segments likely to be fault boundary lines. To solve such problem, the search area for the candidates of the road boundary which is called the region of interest is limited on the Hough space. The effectiveness of the proposed algorithm has been shown with experimental results.

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.143-146
• /
• 2002

The aim of this work is the real-time road edge detection using the fast processing of Cellular Analogic Parallel Processing Networks(CAPPN). The CAPPN is composed of 2D analog cell way. If the dynamic programming is implemented with the CAPPN, the optimal path can be detected in parallel manner Provided that fragments of road edge are utilized as the cost inverse(benefit) in the CAPPN-based optimal path algorithm, the CAPPN determines the most plausible path as the road edge line. Benefits of the proposed algorithm are the fast processing and the utilization of optimal technique to determine the road edge lines.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.8
• /
• pp.773-780
• /
• 2015

Navigation in outdoor environments is a fundamental and challenging problem for unmanned ground vehicles. Detecting lane markings or boundaries on the road may be one of the solutions to make navigation easy. However, because of various environments and road conditions, a robust lane detection is difficult. In this paper, we propose a new approach for estimating virtual lanes on a traversable region. Estimating the virtual lanes consist of two steps: (i) we detect virtual road region through road model selection based on traversability at current frame and similarity between the interframe and (ii) we estimate virtual lane using the number of lane on the road and results of previous frame. To improve the detection performance and reduce the searching region of interests, we use a probability map representing the traversability of the outdoor terrain. In addition, by considering both current and previous frame simultaneously, the proposed method estimate more stable virtual lanes. We evaluate the performance of the proposed approach using real data in outdoor environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.861-880
• /
• 2023

Semantic segmentation of road scene is the key technology of autonomous driving, and the improvement of convolutional neural network architecture promotes the improvement of model segmentation performance. The existing convolutional neural network has the simplification of learning knowledge and the complexity of the model. To address this issue, we proposed a road scene semantic segmentation algorithm based on multi-task collaborative learning. Firstly, a depthwise separable convolution atrous spatial pyramid pooling is proposed to reduce model complexity. Secondly, a collaborative learning framework is proposed involved with saliency detection, and the joint loss function is defined using homoscedastic uncertainty to meet the new learning model. Experiments are conducted on the road and nature scenes datasets. The proposed method achieves 70.94% and 64.90% mIoU on Cityscapes and PASCAL VOC 2012 datasets, respectively. Qualitatively, Compared to methods with excellent performance, the method proposed in this paper has significant advantages in the segmentation of fine targets and boundaries.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.93-101
• /
• 2021

In this paper, a vehicle license plate detection scheme is proposed that uses the spatial attention areas to detect accurately the license plates in various real-road situations. First, the previous WPOD-NET was analyzed, and its detection accuracy is evaluated as lower due to the unnecessary noises in the wide detection candidate areas. To resolve this problem, a vehicle license plate detection model is proposed that uses the candidate area of the license plate as a spatial attention areas. And we compared its performance to that of the WPOD-NET, together with the case of using the optimal spatial attention areas using the ground truth data. The experimental results show that the proposed model has about 20% higher detection accuracy than the original WPOD-NET since the proposed scheme uses tight detection candidate areas.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.9
• /
• pp.4419-4441
• /
• 2016

A cause of vehicle accidents is the reduced visibility due to bad weather conditions such as fog. Therefore, an onboard vision system should take visibility detection into account. In this paper, we propose a simple and effective approach for measuring the visibility distance using a single camera placed onboard a moving vehicle. The proposed algorithm is controlled by a few parameters and mainly includes camera parameter estimation, region of interest (ROI) estimation and visibility computation. Thanks to the ROI extraction, the position of the inflection point may be measured in practice. Thus, combined with the estimated camera parameters, the visibility distance of the input foggy image can be computed with a single camera and just the presence of road and sky in the scene. To assess the accuracy of the proposed approach, a reference target based visibility detection method is also introduced. The comparative study and quantitative evaluation show that the proposed method can obtain good visibility detection results with relatively fast speed.

• Journal of KIISE
• /
• v.43 no.2
• /
• pp.186-195
• /
• 2016

This paper proposes a vehicle license plate detection method in real road environments using 8 bit-MCT features and a landmark-based Adaboost method. The proposed method allows identification of the potential license plate region, and generates a saliency map that presents the license plate's location probability based on the Adaboost classification score. The candidate regions whose scores are higher than the given threshold are chosen from the saliency map. Each candidate region is adjusted by the local image variance and verified by the SVM and the histograms of the 8bit-MCT features. The proposed method achieves a detection accuracy of 85% from various road images in Korea and Europe.