• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.8
• /
• pp.799-806
• /
• 2011

This paper proposes an algorithm to detect the stop line and crosswalk on the road surface using edge information and blob coloring. The detection has been considered as an important area of autonomous vehicle technologies. The proposed algorithm is composed of three phases: 1) hypothesis generation of stop lines, 2) hypothesis generation of crosswalks, and 3) hypothesis verification of stop lines. The last two phases are not performed if the first phase does not provide a hypothesis of a stop line. The last one is carried out by the combination of both hypotheses of stop lines and crosswalks, and determines the stop lines among stop line hypotheses. The proposed algorithm is proven to be effective through experiments with various images captured on the roads.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.1
• /
• pp.35-45
• /
• 2013

A autonomous vehicle requires improved and robust perception systems than conventional perception systems of intelligent vehicles. In particular, single sensor based perception systems have been widely studied by using cameras and laser radar sensors which are the most representative sensors for perception by providing object information such as distance information and object features. The distance information of the laser radar sensor is used for road environment perception of road structures, vehicles, and pedestrians. The image information of the camera is used for visual recognition such as lanes, crosswalks, and traffic signs. However, single sensor based perception systems suffer from false positives and true negatives which are caused by sensor limitations and road environments. Accordingly, information fusion systems are essentially required to ensure the robustness and stability of perception systems in harsh environments. This paper describes a perception system for autonomous vehicles, which performs information fusion to recognize road environments. Particularly, vision and laser radar sensors are fused together to detect lanes, crosswalks, and obstacles. The proposed perception system was validated on various roads and environmental conditions with an autonomous vehicle.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.6
• /
• pp.176-190
• /
• 2019

The prevention of pedestrian accidents in crosswalks and intersections is very important. The C-ITS services provide a warning service for preventing accidents between cars and pedestrians. In the current pedestrian collision prevention warning service according to the C-ITS standard, however, it is difficult to provide real-time service because it detects pedestrians from a video-analysis server in the control center and sends service messages through the ITS system. This paper proposes a pedestrian collision-prevention warning system that detects pedestrians in the local field using an edge camera and sends a warning message directly to the driver through a roadside unit. An evaluation showed that the proposed system could deliver the pedestrian collision prevention-warning message to the driver satisfying the delay time within the 300 ms required by the C-ITS standard, even in the worst case.

• Journal of Korea Multimedia Society
• /
• v.11 no.8
• /
• pp.1051-1058
• /
• 2008

The advancement of technology for image processing and communications makes it possible for current traffic signal controllers and vehicle detection technology to make both emergency vehicle preemption and transit priority strategies as a part of integrated system. Present]y traffic signal control in crosswalk is controlled by fixed signals. The signal control keeps regular signals traffic even with no traffic, when there is traffic, should wait until the signal is given. Waiting time causes the risk of traffic accidents and traffic congestion in accordance with signal violation. To help reduce the risk of accidents and congestion, this paper explains traffic signal control system for the adaptive priority order so that signal may be preferentially given in accordance with the situation of site through the object detect images.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.3
• /
• pp.19-25
• /
• 2013

Recently, more accurate database information of facilities is being required, with the increase in importance of urban road facility management. Therefore, this study proposed how to automatically detect particular traffic signs necessary for efficient construction of road facility DB. For this study, central locations of facilities were searched, after recognition and automatic detection of particular traffic signs through an image. Then, coordinate values of traffic signs calculated in the study were compared with real coordinate values, in order to evaluate the accuracy of traffic sign locations which were finally detected. Computer vision technology was used in recognizing and detecting traffic signs through OPEN CV-based coding, and photogrammetry was used in calculating accurate locations of detected traffic signs. For the experiment, circular road signal(No Parking) and triangular road signal(Crosswalk) were chosen out of various kinds of road signals. The research result showed that the circular road signal had a nearly 50cm error value, and the triangular road signal had a nearly 60cm error value, when comparing the calculated coordinates with the real coordinates. Though this result is not satisfactory, it is considered that there would be no problem to find locations of traffic signs.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.4
• /
• pp.1-9
• /
• 2023

In this paper, we propose a method to recognize the braille blocks for embedded devices in real time through deep learning. First, a deep learning model for braille block recognition is trained on a high-performance computer, and the learning model is applied to a lightweight tool to apply to an embedded device. To recognize the walking information of the braille block, an algorithm is used to determine the path using the distance from the braille block in the image. After detecting braille blocks, bollards, and crosswalks through the YOLOv8 model in the video captured by the embedded device, the walking information is recognized through the braille block path discrimination algorithm. We apply the model lightweight tool to YOLOv8 to detect braille blocks in real time. The precision of YOLOv8 model weights is lowered from the existing 32 bits to 8 bits, and the model is optimized by applying the TensorRT optimization engine. As the result of comparing the lightweight model through the proposed method with the existing model, the path recognition accuracy is 99.05%, which is almost the same as the existing model, but the recognition speed is reduced by 59% compared to the existing model, processing about 15 frames per second.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.166-172
• /
• 2022

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.2
• /
• pp.119-125
• /
• 2023

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.