• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.222-222
• /
• 2019

최근 기후변화로 인한 강우사상의 변화로 가뭄 발생 횟수와 기간이 늘어나는 추세이다. 2013~2018년 전국적으로 장기적인 가뭄이 발생함에 따라 상수도 미 급수 지역에 대한 추가 용수 공급방안을 적용하여 지역 주민의 물 이용 문제를 해결한 바 있다. 장기적으로 물 이용에 대한 갈등이 심화될 것으로 예상되는 가운데 지하수자원의 취약성에 대한 연구가 지속적으로 진행되고 있다. 기존의 연구에서는 주로 지역적인 특성을 반영할 수 있는 매개변수를 설정하고 매개변수 별가중치를 산정하여 공간적인 지하수자원 이용 취약성 평가를 실시하였다. 공간적인 취약성 평가결과는 지하수자원 이용 시기 결정 및 대체 수자원 이용 정책 결정 등 구체적인 대안을 마련하는 근거로서 한계가 있기 때문에 최근 지하수자원 이용에 대한 시간적인 취약성 평가 방법을 개발하는 연구가 진행되고 있다. 본 연구는 보다 구체적인 지하수자원 이용 시기를 결정하기 위하여 금강 유역을 대상으로 분기 별 지하수자원 이용 취약 시기 평가 방법을 개발하였다. 분기 별 지하수자원 이용 취약 시기평가 방법을 개발하기 위하여 우선 연구지역의 지하수위, 하천수위, 강수량 자료를 수집하였다. 수문 관측자료 간의 관계 분석을 통해 물 순환 측면에서의 물리적인 의미를 규명하기 위하여 강수량 자료에 한계침투량 개념을 적용한 강우이동평균 방법을 적용하였고, 하천수위 자료에 대하여 이동평균 방법을 적용하였다. 분기 단위의 지하수자원 이용 취약 시기를 평가함으로써 금강 유역의 지하수자원 이용 취약 시기를 결정하였다. 본 연구는 기존의 공간적인 취약성 평가 방법과 함께 지하수자원 이용 취약성에 대한 시공간적 분석 결과를 제공함으로써 보다 구체적인 지역 별분기 단위 지하수자원 이용 취약 시기를 결정하고, 지역 맞춤형 지하수자원 이용 및 개발 정책에 기여할 수 있을 것으로 기대된다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.9-17
• /
• 2021

Humans are social animals that have gained information or social interaction through dialogue. In conversation, the mood of the word can change depending on the sensibility of one person to another. Relationships between characters in films are essential for understanding stories and lines between characters, but methods to extract this information from films have not been investigated. Therefore, we need a model that automatically analyzes the relationship aspects in the movie. In this paper, we propose a method to analyze the relationship between characters in the movie by utilizing deep learning techniques to measure the emotion of each character pair. The proposed method first extracts main characters from the movie script and finds the dialogue between the main characters. Then, to analyze the relationship between the main characters, it performs a sentiment analysis, weights them according to the positions of the metabolites in the entire time intervals and gathers their scores. Experimental results with real data sets demonstrate that the proposed scheme is able to effectively measure the emotional relationship between the main characters.

• The Journal of the Korea Contents Association
• /
• v.19 no.3
• /
• pp.171-182
• /
• 2019

As online shopping malls continue to grow in popularity, various chances of consumption are provided to customers. Customers decide the purchase by exploiting information provided by shopping malls such as the reviews of actual purchasing users, the detailed information of items, and so on. It is required to provide objective and reliable information because customers have to decide on their own whether the massive information is credible. In this paper, we propose a personalized recommendation method considering an item confidence to recommend reliable items. The proposed method determines user preferences based on various behaviors for personalized recommendation. We also propose an user preference measurement that considers time weights to apply the latest propensity to consume. Finally, we predict the preference score of items that have not been used or purchased before, and we recommend items that have highest scores in terms of both the predicted preference score and the item confidence score.

• Journal of Digital Convergence
• /
• v.17 no.2
• /
• pp.135-144
• /
• 2019

The importance of logistics of warehouse stores has increased as their prices are cheaper and more convenient than those of large supermarkets. However, few studies on integrated depot location selection of warehouse stores have been conducted. In this regard, this study aims to derive factors for integrated depot location selection and calculate weights and select the location priority of target candidates by introducing an analytic hierarchy process (AHP). The analysis results exhibited that the most important selection factor was the cost reduction in transportation and delivery (0.198) followed by distance reduction in transportation and delivery (0.168), and time reduction in transportation. This study quantified the reduction in cost and increase in efficiency if depots were integrated and operated thereby presenting more realistic foundational data to hands-on workers. For the future study, the analysis model will be needed to be advanced through additional investigation on the factors in the analysis.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.1
• /
• pp.79-92
• /
• 2019

In this paper, we propose a method to determine whether to install the accommodation facility of the multi-utility tunnel more effectively in the basic planning phase, and to set up an evaluation system to determine the configuration and kind of accommodation facility. For the configuration of the accommodation facility, 98 alternatives were analyzed for 7 accommodation facility. For the evaluation system of the accommodation facility, index related to feasibility and economic in basic planning phase were selected. The evaluation system classified as spatial, effective, and cost evaluation was presented reflecting the selected index, and AHP analysis was performed for weight setting. The results of this study will be helpful for users including designers to shorten the time and increase the efficiency in the process of determining accommodation facility of the multi-utility tunnel in basic planning phase.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.3
• /
• pp.86-99
• /
• 2021

This research proposes the Accident Prevention System to prevent collision accident that can occur due to blind spots such as crossway or school zone using V2I communication. Vision sensor and LiDAR sensor located in the infrastructure of crossway somewhere like that recognize objects and warn vehicles at risk of accidents to prevent accidents in advance. Using deep learning-based YOLOv4 to recognize the object entering the intersection and using the Manhattan Distance value with LiDAR sensors to calculate the expected collision time and the weight of braking distance and secure safe distance. V2I communication used ROS (Robot Operating System) communication to prevent accidents in advance by conveying various information to the vehicle, including class, distance, and speed of entry objects, in addition to collision warning.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.46-54
• /
• 2022

The existing methods for determining influence in social networks do not accurately determine the influence of users because they do not delete or update existing relationships before they stop in the face of an increasing number of inactive users on social networks. In this paper, we propose a user influence detremination method using the temporal k-shell decomposition technique based on the creation date of users of social networks. To solve the problem of increasing influence of old users in social networks, we apply k-shell decomposition and age-specific order centrality as attenuation coefficients due to aging in neighbors. The age-decaying k-shell decomposition and age-specific order centrality are searched for influential users at the present time by applying the attenuation coefficient and age-dependent weights. Various performance evaluations are performed to show the superiority of the proposed method.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.1-11
• /
• 2021

Slow sync is a photography technique where a user takes an image with long exposure and a camera flash to enlighten the foreground and background. Unlike short exposure with flash and long exposure without flash, slow sync guarantees the bright foreground and background in the dim environment. However, taking a slow sync image with a smartphone is difficult because the smartphone camera has continuous and weak flash and can not turn on flash if the exposure time is long. This paper proposes a deep learning method that input is a short exposure flash image and output is a slow sync image. We present a deep learning network with a weight map for spatially varying enlightenment. We also propose a dataset that consists of smartphone short exposure flash images and slow sync images for supervised learning. We utilize the linearity of a RAW image to synthesize a slow sync image from short exposure flash and long exposure no-flash images. Experimental results show that our method trained with our dataset synthesizes slow sync images effectively.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.83-90
• /
• 2023

Blurred image is an important factor in lowering image recognition rates in Computer vision. This mainly occurs when the camera is unstablely out of focus or the object in the scene moves quickly during the exposure time. Blurred images greatly degrade visual quality, weakening visibility, and this phenomenon occurs frequently despite the continuous development digital camera technology. In this paper, it replace the modified building module based on the Deep multi-patch neural network designed with convolution neural networks to capture details of input images and Attention techniques to focus on objects in blurred images in many ways and strengthen the image. It measures and assigns each weight at different scales to differentiate the blurring of change and restores from rough to fine levels of the image to adjust both global and local region sequentially. Through this method, it show excellent results that recover degraded image quality, extract efficient object detection and features, and complement color constancy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.129-132
• /
• 2022

Super-resolution is the process of converting a low-quality image into a high-quality image. This study was conducted using ESPCN. In a super-resolution deep neural network, different quality images can be output even when receiving the same input data according to the activation function that determines the weight when passing through each node. Therefore, the purpose of this study is to find the most suitable activation function for super-resolution by applying the activation functions ReLU, ELU, and Swish and compare the quality of the output image for the same input images. The CelebaA Dataset was used as the dataset. Images were cut into a square during the pre-processing process then the image quality was lowered. The degraded image was used as the input image and the original image was used for evaluation. As a result, ELU and swish took a long time to train compared to ReLU, which is mainly used for machine learning but showed better performance.