• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1701-1704
• /
• 2005

We developed an machine vision method for navigation control of a traveling automatic guided vehicle(AGV) on desired trajectory with guided marks. The formulated EDF accumulates the edge magnitude for edge directions. The EDF has distinctive peak points at the vicinity of trajectory directions due to the directional and the positional continuities of desired trajectory. Examining the EDF by its shape parameters of the local maxima and symmetry axis results in identifying whether or not change in traveling direction of an AGV has occurred. Simulation results show that the presented method is useful for navigation control of AGV.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.423-425
• /
• 1998

영상검지기를 이용하여 도로상에서 이동중인 차량의 움직임을 탐지하고 분석하는 방법은 지능형교통시스템의 많은 분야에 적용되어질 수 있다. 영상분석으로 움직이는 물체를 탐지하는 방법에는 영상차를 이용하는 방법과 영상차를 이용하지 않는 방법으로 분류할 수 있다. 영상차를 이용하는 방법에서는 영상간의 차영상을 기반으로 하여 물체를 탐지하는 방법은 일반적이고 보편적인 방법이나 시간에 따른 배경영상의 왜곡과 물체의 정체현상에 많은 문제점을 지니고 있다. 그리고 영상차를 이용하지 않는 방법은 영상내의 분석으로 물체를 탐지하는 방법이고, 영상간의 정보를 사용하지 않으므로 영상차에 의한 문제점은 발생되지 않는다. 기존에 연구되어진 영상차를 이용하지 않는 방법은 물체의 형태를 고려하지 않고 단지 이동점의 좌표분석으로 차량의 움직임을 측정하고 있다. 본 논문에서는 영상차를 이용하지 않으며 영상내의 형태정보 분석과 색상정보를 고려하여 기존의 영상검지기가 지니는 문제점을 개선하여 정밀한 차량 추적에 대한 가능성을 알 수 있었다.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2341-2343
• /
• 2002

This paper describes a vision-based line recognition and control of driving direction for an AGV(autonomous guided vehicle). As navigation guide, black stripe attached on the corridor is used. Binary image of guide stripe captured by a CCD camera is used. For detect the guideline quickly and extractly, we use for variable thresholding algorithm. this low-cost line-tracking system is efficiently using pc-based real time vision processing. steering control is studied through controller with guide-line angle error. This method is tested via a typical agv with a single camera in laboratory environment.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.2
• /
• pp.130-136
• /
• 2015

In this paper, we propose a robust curved lane marking detection method. Several lane detection methods have been proposed, however most of them have considered only straight lanes. Compared to the number of straight lane detection researches, less number of curved-lane detection researches has been investigated. This paper proposes a new curved lane detection and tracking method which is robust to various illumination conditions. First, the proposed methods detect straight lanes using a robust road feature image. Using the geometric relation between a vehicle camera and the road plane, several circle models are generated, which are later projected as curved lane models on the camera images. On the top of the detected straight lanes, the curved lane models are superimposed to match with the road feature image. Then, each curve model is voted based on the distribution of road features. Finally, the curve model with highest votes is selected as the true curve model. The performance and efficiency of the proposed algorithm are shown in experimental results.

• International Journal of Highway Engineering
• /
• v.10 no.4
• /
• pp.127-138
• /
• 2008

Vehicles have recently become one of the main factors affecting our quality of life, and the needs of vehicles are still increasing. As a result, the growth of vehicles generate more crashes every year. One main factor for vehicle crashes is uncareful driving behaviors. Especially, drowsy or fatigue driving behaviors explain about 10-20% of the crashes, and they cause serious results because of the delay of response time and the decrease of object-recognition. Therefore, this research conducted real time image processing tests in order to study how cellular phone usages and drowy(or fatigue) drives affect driving behaviors. A vehicle simulator was used for this research, and the faceLAB 4.5 of Seeing Machines for eye image tracking tests using a small camera was installed in the front of the simulator, and normal and drowsy(or fatigue) driving patterns were analyzed.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.4
• /
• pp.689-698
• /
• 2021

In this paper, lane recognition, steering control and speed control algorithms were developed using Bluetooth wireless communication and image processing techniques. Instead of recognizing road traffic signals based on image processing techniques, a methodology for recognizing the permissible road speed by receiving speed codes from electronic traffic signals using Bluetooth wireless communication was developed. In addition, a steering control algorithm based on PWM control that tracks the lanes using the Canny algorithm and Hough transform was developed. A vehicle prototype and a driving test track were developed to prove the accuracy of the developed algorithm. Raspberry Pi and Arduino were applied as main control devices for steering control and speed control, respectively. Also, Python and OpenCV were used as implementation languages. The effectiveness of the proposed methodology was confirmed by demonstrating effectiveness in the lane tracking and driving control evaluation experiments using a vehicle prototypes and a test track.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.6
• /
• pp.619-623
• /
• 2008

A lane detection system is one of the major components of intelligent vehicle systems. Difficulties in lane detection mainly come from not only various weather conditions but also a variety of special environment. This paper describes a simple and stable method for the broken lane tracking in various environments. Probabilistic Hough Transform (PHT) and the Least-square method (LSM) are used to track and correct the lane orientation. For the efficiency of the proposed method, two regions of interest (ROIs) are placed in the lower part of each image, where lane marking areas usually appear with less intervention in our system view. By testing in both a set of static images and video sequences, the experiments showed that the proposed approach yielded robust and reliable results.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.3
• /
• pp.6-13
• /
• 2002

In this paper, we propose an image processing based method to measure red-time and green-time backward moving shockwave speed automatically at signalized intersections. Shockwave means the discontinuous boundary line between different vehicle traffic flows, and its moving speed is called shockwave speed which is obtain from the slope of boundary line. In this paper, we compose distance-time diagram for measuring shockwave speed automatically. By global vehicle tracking, we draw all of the vehicle moving path on distance-time diagram. We analyze the slope change pattern of curved moving path line, and compute red-time and green-time backward moving shockwave speed. We obtain the measurement result of shockwave speed, when applying above mentioned proposed method to experiment at signalized intersections, Once we can measure the shockwave speed, we could apply the result to highway ramp metering and automatic signal control at intersections effectively since we know the situation of frontal congestion easily.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.134-143
• /
• 2020

Vehicle License Plate Recognition is one of the approaches for transportation and traffic safety networks, such as traffic control, speed limit enforcement and runaway vehicle tracking. Although it has been studied for decades, it is attracting more and more attention due to the recent development of deep learning and improved performance. Also, it is largely divided into license plate detection and recognition. In this study, experiments were conducted to improve license plate detection performance by utilizing various object detection methods and WPOD-Net(Warped Planar Object Detection Network) model. The accuracy was improved by selecting the method of detecting the vehicle(s) and then detecting the license plate(s) instead of the conventional method of detecting the license plate using the object detection model. In particular, the final performance was improved through the process of removing noise existing in the image by using the Fast-SRGAN model, one of the Super-Resolution methods. As a result, this experiment showed the performance has improved an average of 4.34% from 92.38% to 96.72% compared to previous studies.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.11
• /
• pp.2277-2282
• /
• 2009

In this paper, we proposed a new recognition method can be used in application systems using morphological features, difference operators and ART2 algorithm. At first, edges are extracted from an acquired car image by a camera using difference operators and the image of extracted edges is binarized by a block binarization method. In order to extract license plate area, noise areas are eliminated by applying morphological features of new and existing types of license plate to the 8-directional edge tracking algorithm in the binarized image. After the extraction of license plate area, mean binarization and mini-max binarization methods are applied to the extracted license plate area in order to eliminated noises by morphological features of individual elements in the license plate area, and then each character is extracted and combined by Labeling algorithm. The extracted and combined characters(letter and number symbols) are recognized after the learning by ART2 algorithm. In order to evaluate the extraction and recognition performances of the proposed method, 200 vehicle license plate images (100 for green type and 100 for white type) are used for experiment, and the experimental results show the proposed method is effective.