• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1329-1341
• /
• 2022

Destroying the marine environment and marine ecosystem and causing marine accidents, marine debris is generated every year, and among them, submerged marine debris is difficult to identify and collect because it is on the seabed. Therefore, deep-learning-based semantic segmentation was experimented on waste fish nets and waste ropes using underwater images to identify efficient collection and distribution. For segmentation, a high-resolution network (HRNet), a state-of-the-art deep learning technique, was used, and the performance of each optimizer was compared. In the segmentation result fish net, F1 score=(86.46%, 86.20%, 85.29%), IoU=(76.15%, 75.74%, 74.36%), For the rope F1 score=(80.49%, 80.48%, 77.86%), IoU=(67.35%, 67.33%, 63.75%) in the order of adaptive moment estimation (Adam), Momentum, and stochastic gradient descent (SGD). Adam's results were the highest in both fish net and rope. Through the research results, the evaluation of segmentation performance for each optimizer and the possibility of segmentation of marine debris in the latest deep learning technique were confirmed. Accordingly, it is judged that by applying the latest deep learning technique to the identification of submerged marine debris through underwater images, it will be helpful in estimating the distribution of marine sedimentation debris through more accurate and efficient identification than identification through the naked eye.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.392-399
• /
• 2023

This paper proposes a deep learning model to determine the region and depth of cabbage cores for robotic automation of the cabbage core removal process during the kimchi manufacturing process. In addition, rather than predicting the depth of the measured cabbage, a model was presented that simultaneously detects and classifies the area by converting it into a discrete class. For deep learning model learning and verification, RGB images of the harvested cabbage 522 were obtained. The core region and depth labeling and data augmentation techniques from the acquired images was processed. MAP, IoU, acuity, sensitivity, specificity, and F1-score were selected to evaluate the performance of the proposed YOLO-v4 deep learning model-based cabbage core area detection and classification model. As a result, the mAP and IoU values were 0.97 and 0.91, respectively, and the acuity and F1-score values were 96.2% and 95.5% for depth classification, respectively. Through the results of this study, it was confirmed that the depth information of cabbage can be classified, and that it can be used in the development of a robot-automation system for the cabbage core removal process in the future.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.534-535
• /
• 2020

본 논문에서는 딥러닝과 지식베이스를 융합하여 활용한 질환 인식 및 식품 추천 시스템을 제안한다. 제안하는 시스템은 온전히 모바일 디바이스 내에서 작동하는 시스템이다. 본 시스템은 압축된 딥러닝 모델을 이용해 사용자 대화 텍스트를 분석하여 사용자의 질환을 예측한다. 그 후, 지식베이스를 기반으로 해당 질환 관리에 도움이 되는 식품을 매칭하고 사용자에게 추천한다. 이는 사용자 친화적 헬스케어 애플리케이션으로써 체크리스트 작성 등 번거로운 작업 없이도 사용자에게 유용한 건강 정보를 제공할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.727-730
• /
• 2024

최근 딥러닝 기술이 급속히 발전하며 현대 사회의 다양한 응용분야에서 빠르게 적용되고 있다. 특히 영상 기반의 딥러닝 기술은 자연어 처리와 함께 인공지능 기술의 핵심 연구 분야로 많은 연구가 진행되고 있다. 논문에서는 최근 많은 연구가 진행되고 있는 영상의 의미적 분할 (Semantic Segmentation) 성능을 향상하기 위한 연구를 진행한다. 특히 모델에서 고정밀의 의미적 분할을 수행할 수 있도록 추가적인 정보로써 의사 깊이맵 (Pseudo Depth-Map)을 활용하는 방법을 제안하였다. 더불어, 의사 깊이맵을 모델 상에서 효과적으로 학습시키기 위하여 다중 디코더 모델과 학습 효율을 높이는 학습 스케줄링 전략을 제안한다. 의사 깊이맵과 다중 디코더 모델 기반의 제안 모델은 기존 의미적 분할 모델과 비교하여 iIoU 기준 2%의 성능 향상을 보였다.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.49 no.1
• /
• pp.85-94
• /
• 2022

This retrospective study aimed to evaluate the difference in measurement between conventional orthodontic analysis and artificial intelligence orthodontic analysis in pediatric and adolescent patients aged 7 - 15 with the mixed and permanent dentition. A total of 60 pediatric and adolescent patients (30 mixed dentition, 30 permanent dentition) who underwent lateral cephalometric radiograph for orthodontic diagnosis were randomly selected. Seventeen cephalometric landmarks were identified, and 22 measurements were calculated by 1 examiner, using both conventional analysis method and deep learning-based analysis method. Errors due to repeated measurements were assessed by Pearson's correlation coefficient. For the mixed dentition group and the permanent dentition group, respectively, a paired t-test was used to evaluate the difference between the 2 methods. The difference between the 2 methods for 8 measurements were statistically significant in mixed dentition group: APDI, SNA, SNB, Mandibular plane angle, LAFH (p < 0.001), Facial ratio (p = 0.001), U1 to SN (p = 0.012), and U1 to A-Pg (p = 0.021). In the permanent dentition group, 4 measurements showed a statistically significant difference between the 2 methods: ODI (p = 0.020), Wits appraisal (p = 0.025), Facial ratio (p = 0.026), and U1 to A-Pg (p = 0.001). Compared with the time-consuming conventional orthodontic analysis, the deep learning-based cephalometric system can be clinically acceptable in terms of reliability and validity. However, it is essential to understand the limitations of the deep learning-based programs for orthodontic analysis of pediatric and adolescent patients and use these programs with the proper assessment.

• The Journal of the Korea Contents Association
• /
• v.14 no.4
• /
• pp.513-521
• /
• 2014

The purpose of this research is to develope and implement a convergent educational contents based on theoretical background of integrated education using agent based modeling in the ubiquitous learning environment. The structure of this contents consists of three modules that were designed by trans-disciplinary concept and situated learning theory. These three modules are: convergent problem presenting module, resource of knowledge module and learning of agent based modeling and IT tools module. After the satisfaction survey of the implemented content, out of 5 total value, the average value was 3.86 for effectiveness, 4.13 for convenience and 3.86 for design. The result of the survey shows that the users are generally satisfied. By using this u-learning contents, learners can experience and learn how to solve the convergent problem by utilizing IT tools without any limitation of device, time and space. At the same time, the proposal of structural design of contents can be a good guideline to the researchers to develop the convergent educational contents in the future.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.10
• /
• pp.15-22
• /
• 2020

In this study, we propose an improvement method that can create U-Net model which detect fine concrete cracks by applying a weighted loss function. Because cracks in concrete are a factor that threatens safety, it is important to periodically check the condition and take prompt initial measures. However, currently, the visual inspection is mainly used in which the inspector directly inspects and evaluates with naked eyes. This has limitations not only in terms of accuracy, but also in terms of cost, time and safety. Accordingly, technologies using deep learning is being researched so that minute cracks generated in concrete structures can be detected quickly and accurately. As a result of attempting crack detection using U-Net in this study, it was confirmed that it could not detect minute cracks. Accordingly, as a result of verifying the performance of the model trained by applying the suggested weighted loss function, a highly reliable value (Accuracy) of 99% or higher and a harmonic average (F1_Score) of 89% to 92% was derived. The performance of the learning improvement plan was verified through the results of accurately and clearly detecting cracks.

• Journal of Advanced Navigation Technology
• /
• v.16 no.1
• /
• pp.76-81
• /
• 2012

With the spread of the various mobile devices, the studies on the learning management system based on the u-learning are actively proceeding. The u-learning-based learning management system is very convenient in that there are no restrictions on the various access devices as well as the access time and place. However, the judgments on the authentication for the user and whether learning is focused on are difficult. In this paper, the voice and user face capture interface rather than the common user event oriented interface was applied to the learning management system. When a user is accessing the learning management system, user's registered password is input and login as voice, and the user's learning attitude is judged through the response utterance of simple words during the process of learning through contents. As a result of evaluating the proposed learning management system, the user's learning achievement and concentration were improved, thus enabling the manager to monitor the user's abnormal learning attitude.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.89-91
• /
• 2022

본 논문에서는 생체인식 수단 중 하나인 손금을 이용한 생체인식에서 Attention U-Net을 기반으로 손금을 추출하는 방법을 제안한다. 손바닥의 손금 중 주요선이라 불리는 생명선, 지능선, 감정선은 거의 변하지 않는 특징을 가지고 있다. 기존의 손금 추출 방법인 비슷한 색상에서 손금 추출, 제한된 Background에서 손금을 추출하는 것이 아닌 피부색과 비슷하거나, 다양한 Background에서 적용될 수 있다. 이를 통해 사용자를 인식하는 생체인식 방법에서 사용할 수 있다. 본 논문에서 사용된 Attention U-Net의 특징을 통해 손금의 Segmentation 영역을 Attention Coefficient를 업데이트하며 효율적으로 학습할 수 있음을 확인하였다.