• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.1996-2015
• /
• 2023

High-definition (HD) maps can provide precise road information that enables an autonomous driving system to effectively navigate a vehicle. Recent research has focused on leveraging semantic segmentation to achieve automatic annotation of HD maps. However, the existing methods suffer from low recognition accuracy in automatic driving scenarios, leading to inefficient annotation processes. In this paper, we propose a novel semantic segmentation method for automatic HD map annotation. Our approach introduces a new encoder, known as the convolutional transformer hybrid encoder, to enhance the model's feature extraction capabilities. Additionally, we propose a multi-level fusion module that enables the model to aggregate different levels of detail and semantic information. Furthermore, we present a novel decoupled boundary joint decoder to improve the model's ability to handle the boundary between categories. To evaluate our method, we conducted experiments using the Bird's Eye View point cloud images dataset and Cityscapes dataset. Comparative analysis against stateof-the-art methods demonstrates that our model achieves the highest performance. Specifically, our model achieves an mIoU of 56.26%, surpassing the results of SegFormer with an mIoU of 1.47%. This innovative promises to significantly enhance the efficiency of HD map automatic annotation.

• Journal of KIISE:Software and Applications
• /
• v.37 no.2
• /
• pp.104-109
• /
• 2010

The lighting reflection is a common problem in image analysis and causes the many difficulties to extract distinct features in related fields. Furthermore, the problem grows in the rainy night. In this paper, we aim to remove light reflection effects and reconstruct a road surface without lighting reflections in order to extract distinct features. The proposed method utilizes a 3D analysis based on a multiple geometry using captured images, with which we can combine each reflected areas; that is, we can remove lighting reflection effects and reconstruct the surface. At first, the regions of lighting sources and reflected surfaces are extracted by local maxima based on vertically projected intensity-histograms. After that, a fundamental matrix and homography matrix among multiple images are calculated by corresponding points in each image. Finally, we combine each surface by selecting minimum value among multiple images and replace it on a target image. The proposed method can reduces lighting reflection effects and the property on the surface is not lost. While the experimental results with collected data shows plausible performance comparing to the speed, reflection-overlapping areas which can not be reconstructed remain in the result. In order to solve this problem, a new reflection model needs to be constructed.

• The KIPS Transactions:PartD
• /
• v.11D no.6
• /
• pp.1259-1268
• /
• 2004

To organize training in limited training area for an actuai combat, realistic training simulation plugged in by various battle conditions is essential. In this paper, we propose a virtual target overlay technique which does not use a virtual image, but Projects a virtual target on ground-based CCD image by appointed scenario for a realistic training simulation. In the proposed method, we create a realistic 3D model (for an instructor) by using high resolution Geographic Tag Image File Format(GeoTIFF) satellite image and Digital Terrain Elevation Data (DTED), and extract the road area from a given CCD image (for both an instructor and a trainee). Satellite images and ground-based sensor images have many differences in observation position, resolution, and scale, thus yielding many difficulties in feature-based matching. Hence, we propose a moving synchronization technique that projects the target on the sensor image according to the marked moving path on 3D satellite image by applying Thin-Plate Spline(TPS) interpolation function, which is an image warping function, on the two given sets of corresponding control point pair. To show the experimental result of the proposed method, we employed two Pentium4 1.8MHz personal computer systems equipped with 512MBs of RAM, and the satellite and sensor images of Daejoen area are also been utilized. The experimental result revealed the effective-ness of proposed algorithm.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1245-1254
• /
• 2023

Significant research has been conducted on the W-band synthetic aperture radar (SAR) system that utilizes the 77 GHz frequency modulation continuous wave (FMCW) radar. To reconstruct the high-resolution W-band SAR image, it is necessary to transform the point cloud acquired from the stereo cameras or the LiDAR in the direction of 6 degrees of freedom (DOF) and apply them to the SAR signal processing. However, there are difficulties in matching images due to the different geometric structures of images acquired from different sensors. In this study, we present the method to extract an optimized depth map by obtaining 6 DOF of the point cloud using a gradient descent method based on the entropy of the SAR image. An experiment was conducted to reconstruct a tree, which is a major road environment object, using the constructed W-band SAR system. The SAR image, reconstructed using the entropy-based gradient descent method, showed a decrease of 53.2828 in mean square error and an increase of 0.5529 in the structural similarity index, compared to SAR images reconstructed from radar coordinates.

• Journal of Digital Convergence
• /
• v.15 no.11
• /
• pp.77-83
• /
• 2017

The Korea Expressway Corporation announced that road casualties on expressways in 2016 were 262 deaths, a 24% decrease compared to 343 deaths in 2015, thanks to the expansion of rest areas for sleepy drivers. And the installation of large-sized banners containing strong messages such as "dozing while driving means your death" helped to reduce the casualty caused by driving while drowsy by 35% compared to that in 2015. Accordingly, this study tried to analyze the impact of public advertisements designed to prohibit dozing while driving on expressways upon drivers, and to present a direction for improvement of such public advertisements in the future. Based on case studies and library researches, the study contemplated the effects of public advertisements on expressways at home and abroad. It was confirmed that the accident rate has been higher on straight roads than on curved roads and that the framing of negative messages using provocative images or slogans on traffic accidents has been considerably effective. In conclusion, if the installation of outdoor billboards for public advertisements at rest areas for sleepy drivers is institutionalized and the systematic provision of information by road section inside and outside of vehicles via Variable Message Sign (VMS) services on expressways, outdoor billboards, or navigation services (including smartphones) is available, it would be possible to maximize the effect of the public advertisements.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.9
• /
• pp.3037-3047
• /
• 2000

This paper describes a method of tracking a close leading vehicle by color image processing using the pairs of tail and brake lights. which emit red light and are housed on the rear of the vehicle in stop-and-go traffic condition. In the color image converted as an HSV color model. candidate regions of rear lights are identified using the color features of a pair of lights. Then. the pair of tailor brake lights are detected by means of the geometrical features and location features for the pattern of the tail and brake lights. The location of the leading vehicle can be estimated by the location of the detected lights and the vehicle can be tracked continuously. It is also possible to detect the braking status of the leading vehicle by measuring the change in HSV color components of the pair of lights detected. In the experiment. this method tracked a leading vehicle successfully from urban road images and was more useful at night than in the daylight. The KAV-Ill (Korea Autonomous Vehicle- Ill) equipped with a color vision system implementing this algorithm was able to follow a leading vehicle autonomously at speeds of up to 15km!h on a paved road at night. This method might be useful for developing an LSA (Low Speed Automation) system that can relieve driver's stress in the stop-and-go traffic conditions encountered on urban roads.

• Journal of Korean Society for Geospatial Information Science
• /
• v.20 no.2
• /
• pp.63-71
• /
• 2012

Recently, web portals have been offering georeferenced SLI(Street-Level Imagery) services, such as Google Streetview. The SLI has a distinctive strength over aerial images or vector maps because it gives us the same view as we see the real world on the street. Based on the characteristic, applicability of the SLI can be increased substantially through conflation with other spatial datasets. However, spatial inconsistency between different datasets is the main reason to decrease the quality of conflation when conflating them. Therefore, this research aims to remove the spatial inconsistency to conflate an SLI with a widely used 2D vector map. The removal of the spatial inconsistency is conducted through three sub-processes of (1) road intersection matching between the SLI trace and the road layer of the vector map for detecting CPPs(Control Point Pairs), (2) inaccurate CPPs filtering by analyzing the trend of the CPPs, and (3) local alignment using accurate CPPs. In addition, we propose an evaluation method suitable for conflation result including an SLI, and verify the effect of the removal of the spatial inconsistency.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.6
• /
• pp.90-100
• /
• 2017

The purpose of this study is to detect and label the vehicles using the drone images as a way to overcome the limitation of the existing point and section detection system and vehicle gap estimation on Arterial road. In order to select the appropriate time zone, position, and altitude for the acquisition of the drone image data, the final image data was acquired by shooting under various conditions. The vehicle was detected by applying mixed Gaussian, image binarization and morphology among various image analysis techniques, and the vehicle was labeled by applying Kalman filter. As a result of the labeling rate analysis, it was confirmed that the vehicle labeling rate is 65% by detecting 185 out of 285 vehicles. The gap was calculated by pixel unitization, and the results were verified through comparison and analysis with Daum maps. As a result, the gap error was less than 5m and the mean error was 1.67m with the preceding vehicle and 1.1m with the following vehicle. The gaps estimated in this study can be used as the density of the urban roads and the criteria for judging the service level.

• Korean Journal of Remote Sensing
• /
• v.20 no.4
• /
• pp.261-274
• /
• 2004

The feature extraction using remotely sensed imagery has been recognized one of the important tasks in remote sensing applications. As the high-resolution imagery are widely used to the engineering purposes, need of more accurate feature information also is increasing. Especially, in case of the automatic extraction of linear feature such as road using mid or low-resolution imagery, several techniques was developed and applied in the mean time. But quantitatively comparative analysis of techniques and case studies for high-resolution imagery is rare. In this study, we implemented a computer program to perform and compare GDPA (Gradient Direction Profile Analysis) algorithm and Hough transformation. Also the results of applying two techniques to some images were compared with road centerline layers and boundary layers of digital map and presented. For quantitative comparison, the ranking method using commission error and omission error was used. As results, Hough transform had high accuracy over 20% on the average. As for execution speed, GDPA shows main advantage over Hough transform. But the accuracy was not remarkable difference between GDPA and Hough transform, when the noise removal was app]ied to the result of GDPA. In conclusion, it is expected that GDPA have more advantage than Hough transform in the application side.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.