• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.4
• /
• pp.77-82
• /
• 2024

Recent advancements in artificial intelligence (AI) technology have led to an increase in the implementation of AI applications in mobile environments. However, due to the limited resources in mobile devices compared to desktops and servers, there is growing interest in research aimed at efficiently executing AI workloads on mobile platforms. While most studies focus on offloading to edge or cloud solutions to mitigate computing resource constraints, research on the characteristics of file I/O related to storage access in mobile settings remains underexplored. This paper analyzes file I/O traces generated during the execution of deep learning applications in mobile environments and investigates how they differ from traditional mobile workloads. We anticipate that the findings of this study will be utilized to design future smartphone system software more efficiently, considering the file access characteristics of deep learning.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.391-401
• /
• 2022

As deep learning-based object detection and recognition research have been developed recently, the scope of application to industry and real life is expanding. But deep learning-based systems in the construction system are still much less studied. Calculating materials in the construction system is still manual, so it is a reality that transactions of wrong volumn calculation are generated due to a lot of time required and difficulty in accurate accumulation. A fast and accurate automatic drawing recognition system is required to solve this problem. Therefore, we propose an AI-based automatic drawing recognition accumulation system that detects and recognizes steel materials in construction drawings. To accurately detect steel materials in construction drawings, we propose data augmentation techniques and spatial attention modules for improving small object detection performance based on YOLOv4. The detected steel material area is recognized by text, and the number of steel materials is integrated based on the predicted characters. Experimental results show that the proposed method increases the accuracy and precision by 1.8% and 16%, respectively, compared with the conventional YOLOv4. As for the proposed method, Precision performance was 0.938. The recall was 1. Average Precision AP_0.5 was 99.4% and AP_0.5:0.95 was 67%. Accuracy for character recognition obtained 99.9.% by configuring and learning a suitable dataset that contains fonts used in construction drawings compared to the 75.6% using the existing dataset. The average time required per image was 0.013 seconds in the detection, 0.65 seconds in character recognition, and 0.16 seconds in the accumulation, resulting in 0.84 seconds.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.7-15
• /
• 2022

Since the direct and indirect damage caused by the fire in the underground utility tunnel will cause great damage to society as a whole, it is necessary to make efforts to prevent and control it in advance. The most of the fires that occur in cables are caused by short circuits, earth leakage, ignition due to over-current, overheating of conductor connections, and ignition due to sparks caused by breakdown of insulators. In order to find the cause of fire at an early stage due to the characteristics of the underground utility tunnel and to prevent disasters and safety accidents, we are constantly managing it with a detection system using image analysis and making efforts. Among them, a case of developing a fire detection system using CCTV-based deep learning image analysis technology has been reported. However, CCTV needs to be supplemented because there are blind spots. Therefore, we would like to develop a high-performance acoustic-based deep learning model that can prevent fire by detecting the spark sound before spark occurs. In this study, we propose a method that can collect sound in underground utility tunnel environments using microphone sensor through development and experiment of prototype module. After arranging an acoustic sensor in the underground utility tunnel with a lot of condensation, it verifies whether data can be collected in real time without malfunction.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.2
• /
• pp.128-133
• /
• 2021

Currently, autonomous vehicle markets are commercializing a third-level autonomous vehicle, but there is a possibility that an accident may occur even during fully autonomous driving due to stability issues. In fact, autonomous vehicles have recorded 81 accidents. This is because, unlike level 3, autonomous vehicles after level 4 have to judge and respond to emergency situations by themselves. Therefore, this paper proposes a vehicle crisis detection system(VCDS) that collects and stores information outside the vehicle through CNN, and uses the stored information and vehicle sensor data to output the crisis situation of the vehicle as a number between 0 and 1. The VCDS consists of two modules. The vehicle external situation collection module collects surrounding vehicle and pedestrian data using a CNN-based neural network model. The vehicle crisis situation determination module detects a crisis situation in the vehicle by using the output of the vehicle external situation collection module and the vehicle internal sensor data. As a result of the experiment, the average operation time of VESCM was 55ms, R-CNN was 74ms, and CNN was 101ms. In particular, R-CNN shows similar computation time to VESCM when the number of pedestrians is small, but it takes more computation time than VESCM as the number of pedestrians increases. On average, VESCM had 25.68% faster computation time than R-CNN and 45.54% faster than CNN, and the accuracy of all three models did not decrease below 80% and showed high accuracy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.236-238
• /
• 2022

According to the current Road Traffic Act, the 2020 amendment bill is currently in effect as a system that designates vehicle types for each lane for the purpose of securing road use efficiency and traffic safety. When comparing the number of traffic accident fatalities per 10,000 vehicles in Germany and Korea, the number of traffic accident deaths in Germany is significantly lower than in Korea. The representative case of the German autobahn, which did not impose a speed limit, suggests that Korea's speeding laws are not the only answer to reducing the accident rate. The designated lane system, which is observed in accordance with the keep right principle of the Autobahn Expressway, plays a major role in reducing traffic accidents. Based on this fact, we propose a traffic enforcement system to crack down on vehicles violating the designated lane system and improve the compliance rate. We develop a designated lane enforcement system that recognizes vehicle types using Yolo5, a deep learning object recognition model, recognizes license plates and lanes using OpenCV, and stores the extracted data in the server to determine whether or not laws are violated.Accordingly, it is expected that there will be an effect of reducing the traffic accident rate through the improvement of driver's awareness and compliance rate.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.89-90
• /
• 2019

본 논문에서는, 에너지 인터넷에서 정밀한 수요반응을 위한 인공지능 알고리즘 모델을 제안한다. 제안하는 인공지능 모델은 시계열 전력사용량 데이터 처리를 위해 딥러닝 기반 long-short term memory (LSTM) 네트워크를 사용한다. 시뮬레이션 결과를 통해 제안한 시스템 모델의 전력사용량 예측 정확도를 확인하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.99-101
• /
• 2017

최근 딥러닝을 이용한 SAR 영상의 목표물을 인식하는 알고리즘이 괄목할만한 성능을 보여주었다. 이러한 알고리즘들은 포즈 각도 정보를 무시한 채 목표물의 종류를 추정하는 것에만 초점을 맞춘다. 포즈 각도 추정 알고리즘은 단지 SAR 영상 목표물 인식 알고리즘의 전처리 과정으로 연구되었다. 하지만 감시 시스템에서, 목표물이 향하고 있는 방향을 추정하는 것 또한 중요하다. 먼저, 포즈 각도 추정을 통하여 적의 전술 배치를 계획을 추정할 수 있다. 또한 목표물이 아군 쪽을 바라보면 큰 위협이 되는데, 포즈 각도 추정을 통하여 이러한 정보를 알 수 있다. 따라서 본 논문은 목표물이 향하고 방향을 추정할 수 있는 콘볼루션 네트워크를 고안하였다. 네트워크를 학습시키기 위하여 SAR 영상의 목표물의 포즈 각도를 양자화하여 포즈 각도 label 을 구성하였다. 또한 이러한 포즈 각도 추정을 정제하는 알고리즘을 고안하였고 이는 보다 정확한 포즈 각도 추정을 가능하게 하였다. 그 결과, 제안된 네트워크는 포즈 각도 추정에 높은 정확도를 보여준다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.7
• /
• pp.928-933
• /
• 2018

Nowadays, gastropathy is a common disease. As endoscopic equipment are developed and used widely, it is possible to provide a large number of endoscopy images. Computer-aided Diagnosis (CADx) systems aim at helping physicians to identify possibly malignant abnormalities more accurately. In this paper, we present a CADx system to detect and classify the abnormalities of gastric lesions which include bleeding, ulcer, neuroendocrine tumor and cancer. We used an Inception module based deep learning model. And we used data augmentation for learning. Our preliminary results demonstrated promising potential for automatically labeled region of interest for endoscopy doctors to focus on abnormal lesions for subsequent targeted biopsy, with Az values of Receiver Operating Characteristic(ROC) curve was 0.83. The proposed CADx system showed reliable performance.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.570-573
• /
• 2018

본 논문에서는 한국 가요 학습 데이터를 노래 가사 마디 단위로 문자열을 역전시키는 형태로 변형하고 LSTM으로 학습하여, 마디 간의 문맥을 고려해 문자열을 생성하는 방법에 대해 제안한다. 그리고 이를 통해 특정 가요 가사와 유사하면서도 다른 가사를 생성하는 것도 가능하다. 문자열의 우측 끝에 위치하면서 마디 간의 문맥을 연결해 주는 서술어, 접속사와 같은 요소를 활용하기 위해 데이터를 변형하여 적용한다. 제안하는 방식으로 생성한 문자열이 단순히 문자열 데이터를 그대로 학습하여 생성하는 것보다 상대적으로 더 자연스러운 문맥으로 생성되는 것을 확인하였다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.56-61
• /
• 2016

본 논문에서는 Sequence-to-sequence 모델을 생성요약의 방법으로 한국어 문서요약에 적용하였으며, copy mechanism과 input feeding을 적용한 RNN search 모델을 사용하여 시스템의 성능을 높였다. 인터넷 신문기사를 수집하여 구축한 한국어 문서요약 데이터 셋(train set 30291 문서, development set 3786 문서, test set 3705문서)으로 실험한 결과, input feeding과 copy mechanism을 포함한 모델이 형태소 기준으로 ROUGE-1 35.92, ROUGE-2 15.37, ROUGE-L 29.45로 가장 높은 성능을 보였다.