• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.4
• /
• pp.445-450
• /
• 2018

AIS is the most important navigation equipment for the identification of other ships, etc. However, the AIS overload problem has been raised recently due to an increase in AIS equipped vessels. The government is planning to introduce the wireless LTE network at 100 km offshore as part of the SMART-Navigation project. Continuous development and dissemination of the services available through such platforms will be necessary to achieve major goals such as marine accident prevention and environmental protection. In this study, we applied a WAVE communication system, which could be the basis for the development of such services. As a result, reliable data transmission was confirmed for a range of communication of approx. 5 miles, although the service was limited to 1 km in road traffic. Therefore, it is expected that WAVE communication technology will be used to prevent marine accidents through such efforts as collision avoidance and the transfer of marine safety information between ships.

• Journal of Korea Multimedia Society
• /
• v.22 no.1
• /
• pp.44-54
• /
• 2019

Recently, image-based object detection has made great progress with the introduction of Convolutional Neural Network (CNN). Many trials such as Region-based CNN, Fast R-CNN, and Faster R-CNN, have been proposed for achieving better performance in object detection. YOLO has showed the best performance under consideration of both accuracy and computational complexity. However, these data-driven detection methods including YOLO have the fundamental problem is that they can not guarantee the good performance without a large number of training database. In this paper, we propose a data sampling method using CycleGAN to solve this problem, which can convert styles while retaining the characteristics of a given input image. We will generate the insufficient data samples for training more robust object detection without efforts of collecting more database. We make extensive experimental results using the day-time and night-time road images and we validate the proposed method can improve the object detection accuracy of the night-time without training night-time object databases, because we converts the day-time training images into the synthesized night-time images and we train the detection model with the real day-time images and the synthesized night-time images.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.63-74
• /
• 2021

Location-based services (LBSs) are expected to process a large number of spatial queries, such as shortest path and k-nearest neighbor queries that arrive simultaneously at peak periods. Deploying more LBS servers to process these simultaneous spatial queries is a potential solution. However, this significantly increases service operating costs. Recently, batch processing solutions have been proposed to process a set of queries using shareable computation. In this study, we investigate the problem of batch processing moving k-nearest neighbor (MkNN) queries in dynamic spatial networks, where the travel time of each road segment changes frequently based on the traffic conditions. LBS servers based on one-query-at-a-time processing often fail to process simultaneous MkNN queries because of the significant number of redundant computations. We aim to improve the efficiency algorithmically by processing MkNN queries in batches and reusing sharable computations. Extensive evaluation using real-world roadmaps shows the superiority of our solution compared with state-of-the-art methods.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.29-35
• /
• 2021

License plate recognition plays a key role in intelligent transportation systems. Therefore, it is a very important process to efficiently detect the number and character areas. In this paper, we propose a method to effectively detect license plate number area by applying deep learning and semantic image segmentation algorithm. The proposed method is an algorithm that detects number and text areas directly from the license plate without preprocessing such as pixel projection. The license plate image was acquired from a fixed camera installed on the road, and was used in various real situations taking into account both weather and lighting changes. The input images was normalized to reduce the color change, and the deep learning neural networks used in the experiment were Vgg16, Vgg19, ResNet18, and ResNet50. To examine the performance of the proposed method, we experimented with 500 license plate images. 300 sheets were used for learning and 200 sheets were used for testing. As a result of computer simulation, it was the best when using ResNet50, and 95.77% accuracy was obtained.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.1
• /
• pp.146-159
• /
• 2021

vehicle simulator as part of an empirical analysis the driving characteristics of elderly drivers. To this end, the driving characteristics of the elderly driver from previous study review. he driving characteristics of the elderly the driving elderly driver and general driverIn summarizing these experimental results, the -test showed different driving characteristics from general drivers in all items except for one side of the lane, such as driving speed and driving operation (brake, throttle, steering operation) at a significance level of 95%. Second, when changing lanes, it was difficult for elderly driver to maintain speed and secure an appropriate distance between carslderly driver changed lanes even in inappropriate situations (short distances between cars). Third, in unexpected situation, elderly drivers needed more distance and time.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.4
• /
• pp.46-56
• /
• 2021

Predicting the incident clearance time is important for eliminating the high transportation costs and congestion from non-repetitive congestion caused by incidents. In this study, the factors influencing the clearance time suitable for domestic road conditions were analyzed, using a training dataset for predicting the incident clearance time using artificial neural networks. In a previous study, the under-prediction problem for high incident clearance time was used. In the present study, over-sampling training data applied using the SMOGN technique was obtained and applied to the model as a solution. As a result, the DNN model applying the SMOGN technique could compensate for the limitations of the previously developed prediction model by predicting the clearance time with the highest accuracy among the models developed in the research process with MAE = 18.3 minutes.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.6
• /
• pp.827-833
• /
• 2022

This study aims to prevent secondary traffic accidents with high severity by overcoming the limitations of existing traffic information collection systems through analysis of traffic information collection detectors and various algorithms used to detect unexpected situations. In other words, this study is meaningful present that analyzing the 'unexpected situation that causes secondary traffic accidents' and 'Existing traffic information collection system' accordingly presenting a solution that can preemptively prevent secondary traffic accidents, intelligent traffic information collection system that enables accurate information collection on all sections of the road. As a result of the experiment, the reliability of data transmission reached 97% based on 95%, the data transmission speed averaged 209ms based on 1000ms, and the network failover time achieved targets of 50sec based on 120sec.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3904-3922
• /
• 2022

As a research hotspot, pedestrian detection has a wide range of applications in the field of computer vision in recent years. However, current pedestrian detection methods have problems such as insufficient detection accuracy and large models that are not suitable for large-scale deployment. In view of these problems mentioned above, a lightweight pedestrian detection and early warning method using a new model called you only look once (Yolov5) is proposed in this paper, which utilizing advantages of Yolov5s model to achieve accurate and fast pedestrian recognition. In addition, this paper also optimizes the loss function of the batch normalization (BN) layer. After sparsification, pruning and fine-tuning, got a lot of optimization, the size of the model on the edge of the computing power is lower equipment can be deployed. Finally, from the experimental data presented in this paper, under the training of the road pedestrian dataset that we collected and processed independently, the Yolov5s model has certain advantages in terms of precision and other indicators compared with traditional single shot multiBox detector (SSD) model and fast region-convolutional neural network (Fast R-CNN) model. After pruning and lightweight, the size of training model is greatly reduced without a significant reduction in accuracy, and the final precision reaches 87%, while the model size is reduced to 7,723 KB.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.2
• /
• pp.321-327
• /
• 2023

With the development of artificial intelligence, the prediction system has become one of the essential technologies in our lives. Despite the growth of these technologies, traffic congestion at intersections in the 21st century has continued to be a problem. This paper proposes a system that predicts intersection traffic jams using a Convolutional LSTM (Conv-LSTM) algorithm. The proposed system models data obtained by learning traffic information by time zone at the intersection where traffic congestion occurs. Traffic congestion is predicted with traffic volume data recorded over time. Based on the predicted result, the intersection traffic signal is controlled and maintained at a constant traffic volume. Road congestion data was defined using VDS sensors, and each intersection was configured with a Conv-LSTM algorithm-based network system to facilitate traffic.

• Journal of The Geomorphological Association of Korea
• /
• v.27 no.2
• /
• pp.81-97
• /
• 2020

Geopark refers to a single boundary area consisting of a collection of geosites and geotrails, which includes ecological, historical and cultural elements based on geological and geomorphological resources. To ensure the continued development and conservation of existing listed geoparks, it is necessary to carry out an environmental sensitivity analysis of the geopark components by utilizing spatial information from various scales. The objectives of this study are to analyze the environmental sensitivity in Cheongsong UNESCO global geopark in relation with slope disaster using GIS and to understand its spatial distribution in connection with geosites and geotrails. Two types of spatial database were constructed; geosites and geotrails in Cheongsong UNESCO global geopark and spatial data to perform environmental sensitivity. Potential soil loss and slope stability were analyzed to derive environmental sensitivity related to slope hazard. The results showed relatively high environmental sensitivity along the drainage network of Cheongsong UNESCO global geopark. Zonal statistics analysis was conducted for further detailed distribution of environmental sensitivity based on buffer zones of geosites and geotrails. Majority of geological sites, geological trails, Jeolgol gorge~Jusan Pond section in hiking trails, and Dalgi Mineral Spring Site~Artistic Genius Republic of Korea(Jangnankki gonghwaguk) section in road areas show relatively high slope hazard sensitivity within buffer zones.