• 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.

• Structural Engineering and Mechanics
• /
• v.91 no.5
• /
• pp.443-457
• /
• 2024

At present, the traditional concrete surface inspection methods based on artificial vision have the problems of high cost and insecurity, while the computer vision methods rely on artificial selection features in the case of sensitive environmental changes and difficult promotion. In order to solve these problems, this paper introduces deep learning technology in the field of computer vision to achieve automatic feature extraction of structural damage, with excellent detection speed and strong generalization ability. The main contents of this study are as follows: (1) A method based on DeepLabV3+ convolutional neural network model is proposed for surface detection of post-earthquake structural damage, including surface damage such as concrete cracks, spaling and exposed steel bars. The key semantic information is extracted by different backbone networks, and the data sets containing various surface damage are trained, tested and evaluated. The intersection ratios of 54.4%, 44.2%, and 89.9% in the test set demonstrate the network's capability to accurately identify different types of structural surface damages in pixel-level segmentation, highlighting its effectiveness in varied testing scenarios. (2) A semantic segmentation model based on DeepLabV3+ convolutional neural network is proposed for the detection and evaluation of post-earthquake structural components. Using a dataset that includes building structural components and their damage degrees for training, testing, and evaluation, semantic segmentation detection accuracies were recorded at 98.5% and 56.9%. To provide a comprehensive assessment that considers both false positives and false negatives, the Mean Intersection over Union (Mean IoU) was employed as the primary evaluation metric. This choice ensures that the network's performance in detecting and evaluating pixel-level damage in post-earthquake structural components is evaluated uniformly across all experiments. By incorporating deep learning technology, this study not only offers an innovative solution for accurately identifying post-earthquake damage in civil engineering structures but also contributes significantly to empirical research in automated detection and evaluation within the field of structural health monitoring.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1726-1748
• /
• 2024

Trajectory planning is vital for autonomous systems like robotics and UAVs, as it determines optimal, safe paths considering physical limitations, environmental factors, and agent interactions. Recent advancements in trajectory planning and future location prediction stem from rapid progress in machine learning and optimization algorithms. In this paper, we proposed a novel framework for Spatial-temporal transformer-based feed-forward neural networks (STTFFNs). From the traffic flow local area point of view, skip-gram model is trained on trajectory data to generate embeddings that capture the high-level features of different trajectories. These embeddings can then be used as input to a transformer-based trajectory planning model, which can generate trajectories for new objects based on the embeddings of similar trajectories in the training data. In the next step, distant regions, we embedded feedforward network is responsible for generating the distant trajectories by taking as input a set of features that represent the object's current state and historical data. One advantage of using feedforward networks for distant trajectory planning is their ability to capture long-term dependencies in the data. In the final step of forecasting for future locations, the encoder and decoder are crucial parts of the proposed technique. Spatial destinations are encoded utilizing location-based social networks(LBSN) based on visiting semantic locations. The model has been specially trained to forecast future locations using precise longitude and latitude values. Following rigorous testing on two real-world datasets, Porto and Manhattan, it was discovered that the model outperformed a prediction accuracy of 8.7% previous state-of-the-art methods.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.109-123
• /
• 2024

Research into vision-based end-to-end autonomous driving systems utilizing deep learning and reinforcement learning has been steadily increasing. These systems typically encode continuous and high-dimensional vehicle states, such as location, velocity, orientation, and sensor data, into latent features, which are then decoded into a vehicular control policy. The complexity of urban driving environments necessitates the use of state representation learning through networks like Variational Autoencoders (VAEs) or Convolutional Neural Networks (CNNs). This paper analyzes the impact of different image state encoding methods on reinforcement learning performance in autonomous driving. Experiments were conducted in the CARLA simulator using RGB images and semantically segmented images captured by the vehicle's front camera. These images were encoded using VAE and Vision Transformer (ViT) networks. The study examines how these networks influence the agents' learning outcomes and experimentally demonstrates the role of each state representation technique in enhancing the learning efficiency and decision- making capabilities of autonomous driving systems.

• ETRI Journal
• /
• v.29 no.3
• /
• pp.281-291
• /
• 2007

Wearable computers consisting of various small devices such as smart phones, digital cameras, MP3 players and specialized I/O devices in personal area networks will play an important role in future ubiquitous computing. In this environment, accessing user data is quite complex due to the dynamic and heterogeneous characteristics of the underlying networks. Moreover, since the amount of user data increases rapidly, automatic data backup management is also critical. To overcome these challenges, several studies have been conducted including our previously proposed file service system, PosCFS, which could be adapted to the requirements with a virtualization technique allowing per-user global namespace for managing and accessing data stored on physical storage spaces detected in PAN. In this paper, we present a smart file service framework, PosCFS+ which is an improved and extended version of our previous work. Performance improvement is made possible by redesigning the metadata management scheme based on database and keywords rather than ontology. In addition, the automatic data replication management is newly designed based on the OSD protocol.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• Journal of Contemporary Eastern Asia
• /
• v.20 no.2
• /
• pp.85-107
• /
• 2021

Wildfires have become increasingly common and intense in South Korea because of climate change, but few have recognized the catastrophic level of the problem. Given the significant impact of wildfires, emergency management stakeholders must have effective risk communication structures for rapidly responding to such phenomena and overcoming geographical difficulties. Despite the country spending billions of dollars to build a big databased early warning system, risk communication flow during the 2017 Gangneung wildfire was ineffective, thereby causing substantial economic, social, and environmental losses. To examine the patterns of information exchange in South Korea's risk communication networks and their structural characteristics during the wildfire, we conducted semantic and network analyses of real-time data collected from social media. The results showed that the inefficient flow of risk information prevented emergency responders from adequately assessing the emergency and protecting the population. This study provides new insights into effective risk communication responses to catastrophic events and methods of research on webometric approaches to emergency management.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3230-3255
• /
• 2022

Causality mining in NLP is a significant area of interest, which benefits in many daily life applications, including decision making, business risk management, question answering, future event prediction, scenario generation, and information retrieval. Mining those causalities was a challenging and open problem for the prior non-statistical and statistical techniques using web sources that required hand-crafted linguistics patterns for feature engineering, which were subject to domain knowledge and required much human effort. Those studies overlooked implicit, ambiguous, and heterogeneous causality and focused on explicit causality mining. In contrast to statistical and non-statistical approaches, we present Bidirectional Encoder Representations from Transformers (BERT) integrated with Multi-level Feature Networks (MFN) for causality recognition, called BERT+MFN for causality recognition in noisy and informal web datasets without human-designed features. In our model, MFN consists of a three-column knowledge-oriented network (TC-KN), bi-LSTM, and Relation Network (RN) that mine causality information at the segment level. BERT captures semantic features at the word level. We perform experiments on Alternative Lexicalization (AltLexes) datasets. The experimental outcomes show that our model outperforms baseline causality and text mining techniques.

• Journal of Internet Computing and Services
• /
• v.24 no.1
• /
• pp.71-86
• /
• 2023

The purpose of this exploratory study is to compare the differences in public perceptual characteristics of Korean and Chinese societies regarding the metaverse using big data analysis. Due to the environmental impact of the COVID-19 pandemic, technological progress, and the expansion of new consumer bases such as generation Z and Alpha, the world's interest in the metaverse is drawing attention, and related academic studies have been also in full swing from 2021. In particular, Korea and China have emerged as major leading countries in the metaverse industry. It is a timely research question to discover the difference in social awareness using big data accumulated in both countries at a time when the amount of mentions on the metaverse has skyrocketed. The analysis technique identifies the importance of key words by analyzing word frequency, N-gram, and TF-IDF of clean data through text mining analysis, and analyzes the density and centrality of semantic networks to determine the strength of connection between words and their semantic relevance. Python 3.9 Anaconda data science platform 3 and Textom 6 versions were used, and UCINET 6.759 analysis and visualization were performed for semantic network analysis and structural CONCOR analysis. As a result, four blocks, each of which are similar word groups, were driven. These blocks represent different perspectives that reflect the types of social perceptions of the metaverse in both countries. Studies on the metaverse are increasing, but studies on comparative research approaches between countries from a cross-cultural aspect have not yet been conducted. At this point, as a preceding study, this study will be able to provide theoretical grounds and meaningful insights to future studies.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1693-1705
• /
• 2023

Rural areas, which account for about 90% of the country's land area, are increasing in importance and value as a space that performs various public functions. However, facilities that adversely affect residents' lives, such as livestock facilities, factories, and solar panels, are being built indiscriminately near residential areas, damaging the rural environment and landscape and lowering the quality of residents' lives. In order to prevent disorderly development in rural areas and manage rural space in a planned manner, detection and monitoring of hazardous facilities in rural areas is necessary. Data can be acquired through satellite imagery, which can be acquired periodically and provide information on the entire region. Effective detection is possible by utilizing image-based deep learning techniques using convolutional neural networks. Therefore, U-Net model, which shows high performance in semantic segmentation, was used to classify potentially hazardous facilities in rural areas. In this study, KOMPSAT ortho-mosaic optical imagery provided by the Korea Aerospace Research Institute in 2020 with a spatial resolution of 0.7 meters was used, and AI training data for livestock facilities, factories, and solar panels were produced by hand for training and inference. After training with U-Net, pixel accuracy of 0.9739 and mean Intersection over Union (mIoU) of 0.7025 were achieved. The results of this study can be used for monitoring hazardous facilities in rural areas and are expected to be used as basis for rural planning.