• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.519-520
• /
• 2023

본 논문에서는 최근 반려동물 등록제가 확대되고 있는 바, 기존의 마이크로 칩 삽입 방법을 회피하고 반려견 이미지를 통하여 개체를 인식하는 방법을 연구하였다. 반려견의 전체 이미지를 학습시켜 해당 개체를 식별하는 지능형 시스템을 ResNet 알고리즘을 이용하여 구현하고, 수집된 반려견의 개체 사진을 학습시켜 필요한 개체를 식별할 수 있도록 하였다.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.724-726
• /
• 2023

컴퓨터 비전 문제 중 이미지 분류는 핵심적인 주제 중 하나이다. 딥러닝의 발전으로 이미지 분류 문제에서 높은 정확도와 성능을 보여준다. 하지만 대부분 이미지 분류 연구에서 시각정보인 이미지 내의 특징에만 의존하고 있다. 그렇기에 이미지의 본질적인 맥략과 함께 있는 텍스트 정보를 활용하지 못하는 경우도 있다. 이에 본 논문은 텍스트 정보를 활용하여 이미지 분류 성능을 개선하는 방식을 제안한다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.5
• /
• pp.1001-1010
• /
• 2021

Normal people are not deeply aware of their dependence on sight when eating. However, since the visually impaired do not know what kind of food is on the table, the assistant next to them holds the blind spoon and explains the position of the food in a clockwise direction, front and rear, left and right, etc. In this paper, we describe the development of a meal assistance system that recognizes each food image and announces the name of the food by voice when a visually impaired person looks at their table using a smartphone camera. This system extracts the food on which the spoon is placed through the YOLO model that has learned the image of food and tableware (spoon), recognizes what the food is, and notifies it by voice. Through this system, it is expected that the visually impaired will be able to eat without the help of a meal assistant, thereby increasing their self-reliance and satisfaction.

• Journal of IKEEE
• /
• v.23 no.3
• /
• pp.777-784
• /
• 2019

A pre-fire awareness and automatic notification system are required because it is possible to minimize the damage if the fire situation is precisely detected after a fire occurs in a place where people are unusual or in a mountainous area. In this study, we developed a RaspberryPi-based fire recognition system using Faster-recurrent convolutional neural network (F-RCNN) and single shot multibox detector (SSD) and demonstrated a fire alarm system that works with power line communication. Image recognition was performed with a pie camera of RaspberryPi, and the detected fire image was transmitted to a monitoring PC through an inductive power line communication network. The frame rate per second (fps) for each learning model was 0.05 fps for Faster-RCNN and 1.4 fps for SSD. SSD was 28 times faster than F-RCNN.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.4
• /
• pp.125-144
• /
• 2022

When a variable message signs (VMS) system displays false information related to traffic safety caused by malicious attacks, it could pose a serious risk to drivers. If the normal message patterns displayed on the VMS system are learned, it would be possible to detect and respond to the anomalous messages quickly. This paper proposes a method for detecting anomalous messages by learning the normal patterns of messages using a bi-directional generative pre-trained transformer (GPT) network. In particular, the proposed method was trained using the normal messages and their system parameters to minimize the corresponding negative log-likelihood (NLL) values. After adequate training, the proposed method could detect an anomalous message when its NLL value was larger than a pre-specified threshold value. The experiment results showed that the proposed method could detect malicious messages and cases when the system error occurs.

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we propose an autonomous driving system using an end-to-end model to improve lane departure and misrecognition of traffic lights in a vision sensor-based system. End-to-end learning can be extended to a variety of environmental conditions. Driving data is collected using a model car based on a vision sensor. Using the collected data, it is composed of existing data and data with outlayers removed. A class was formed with camera image data as input data and speed and steering data as output data, and data learning was performed using an end-to-end model. The reliability of the trained model was verified. Apply the learned end-to-end model to the model car to predict the steering angle with image data. As a result of the learning of the model car, it can be seen that the model with the outlayer removed is improved than the existing model.

• Smart Media Journal
• /
• v.12 no.4
• /
• pp.27-40
• /
• 2023

Many researchers have investigated ways to enhance recommender systems by integrating heterogeneous data to address the data sparsity problem. However, only a few studies have successfully integrated heterogeneous data using knowledge graph. Additionally, most of the knowledge graphs built in these studies only incorporate explicit relationships between entities and lack additional information. Therefore, we propose a method for expanding knowledge graphs by using deep learning to model latent relationships between heterogeneous data from multiple knowledge bases. Our extended knowledge graph enhances the quality of entity features and ultimately increases the accuracy of predicted user preferences. Experiments using real music data demonstrate that the expanded knowledge graph leads to an increase in recommendation accuracy when compared to the original knowledge graph.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.3
• /
• pp.469-475
• /
• 2022

To train the model of the deep learning-based single-speaker TTS system, a speech DB of tens of hours and a lot of training time are required. This is an inefficient method in terms of time and cost to train multi-speaker or personalized TTS models. The voice cloning method uses a speaker encoder model to make the TTS model of a new speaker. Through the trained speaker encoder model, a speaker embedding vector representing the timbre of the new speaker is created from the small speech data of the new speaker that is not used for training. In this paper, we propose a multi-speaker TTS system to which voice cloning is applied. The proposed TTS system consists of a speaker encoder, synthesizer and vocoder. The speaker encoder applies the d-vector technique used in the speaker recognition field. The timbre of the new speaker is expressed by adding the d-vector derived from the trained speaker encoder as an input to the synthesizer. It can be seen that the performance of the proposed TTS system is excellent from the experimental results derived by the MOS and timbre similarity listening tests.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.357-359
• /
• 2021

Railway system as a mass transportation is under progress to smart railway system beyond high speed and automation era. Communication network technology including 5G-R(5th Generation - Railway) mobile communication technology and information convergence technology of Big Data, Deep Learnig, AI(Artificial Intelliegnce) and Block Chain have to be used for implementation and operation of this smart railway system. In this paper, a communication network structure is suggested for this smart railway system. This suggested smart railway commnuication network structure is composed with layered structure of plane unit for safety operation of high speed railway, railway system management and customer services, and also have some complexed function of each plane. Results of this study can be used for smart railway communication network implementation, operation and managements, development of railway communication standards.

• The Journal of Bigdata
• /
• v.7 no.1
• /
• pp.63-73
• /
• 2022

COVID-19 has continued from 2020 to the present, and many social changes have occurred. Wearing a mask has become mandatory, and if you do not wear a mask, you cannot use public facilities or restaurants. For this reason, most public facility entrances are equipped with a mask recognition system to check whether a mask is worn. However, it is unclear whether people who cover their mouths with a scarf or who do not wear a mask properly can be identified. In this study, we proposed an embedded mask recognition system using YOLOv5. Unlike the existing mask recognition system, it was able to distinguish not only whether a mask was worn, but also whether a mask was worn in various exceptional situations, such as a person with a scarf or a person covering their mouth with their hands, and showed excellent performance when mounted on the Nvida Jetson Nano Board.