• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.71-80
• /
• 2024

In this study, a proposed ensemble learning technique aims to enhance the semantic segmentation performance of images captured by Unmanned Aerial Vehicles (UAVs). With the increasing use of UAVs in fields such as urban planning, there has been active development of techniques utilizing deep learning segmentation methods for land cover segmentation. The study suggests a method that utilizes prominent segmentation models, namely U-Net, DeepLabV3, and Fully Convolutional Network (FCN), to improve segmentation prediction performance. The proposed approach integrates training loss, validation accuracy, and class score of the three segmentation models to enhance overall prediction performance. The method was applied and evaluated on a land cover segmentation problem involving seven classes: buildings,roads, parking lots, fields, trees, empty spaces, and areas with unspecified labels, using images captured by UAVs. The performance of the ensemble model was evaluated by mean Intersection over Union (mIoU), and the results of comparing the proposed ensemble model with the three existing segmentation methods showed that mIoU performance was improved. Consequently, the study confirms that the proposed technique can enhance the performance of semantic segmentation models.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1371-1388
• /
• 2023

Floods are becoming more severe and frequent due to global warming-induced climate change. Water disasters are rising in Korea due to severe rainfall and wet seasons. This makes preventive climate change measures and efficient water catastrophe responses crucial, and synthetic aperture radar satellite imagery can help. This research created 1,423 water body learning datasets for individual water body regions along the Han and Nakdong waterways to reflect domestic water body properties discovered by Sentinel-1 satellite radar imagery. We created a document with exact data annotation criteria for many situations. After the dataset was processed, U-Net, a deep learning model, analyzed water body detection results. The results from applying the learned model to water body locations not involved in the learning process were studied to validate soil water body monitoring on a national scale. The analysis showed that the created water body area detected water bodies accurately (F1-Score: 0.987, Intersection over Union [IoU]: 0.955). Other domestic water body regions not used for training and evaluation showed similar accuracy (F1-Score: 0.941, IoU: 0.89). Both outcomes showed that the computer accurately spotted water bodies in most areas, however tiny streams and gloomy areas had problems. This work should improve water resource change and disaster damage surveillance. Future studies will likely include more water body attribute datasets. Such databases could help manage and monitor water bodies nationwide and shed light on misclassified regions.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.231-237
• /
• 2020

In this paper, we implemented a home IoT system based on MQTT protocol. In this system, data are collected from sensors in real time and transmitted to the server system. Based on collected data, home devices could be controlled automatically or manually. By using the MQTT protocol, we were able to see the data values of sensors collected in real time according to the topic setting. We implemented a system that automatically sets up home devices based on topic data, and it worked. The system is expected to be useful in applications that require monitoring and tracking of data in real time.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.01a
• /
• pp.93-94
• /
• 2015

최근 산업 현장에서 재해는 설비의 다양화, 인적 구성의 복잡성, 작업환경의 변화 등으로 다양하게 발생되고 있다. 특히 경제적 자립도가 취약한 중소기업에서는 관리능력이 미흡하여 안전에 대한 조직 및 관리, 교육 등이 큰 문제점으로 대두되고 있다. 우리나라의 안전관리의 현실은 대기업 중심으로 이루어지고 있는 실정이나 재해율은 일본 등 주변국보다 2~4배 높은 수준으로 선진국에 비해 아직도 매우 심각한 문제이다. 한편, 환경과 재난방지, 헬스케어 등과 같은 분야에 널리 활용되고 있는 고도화된 IoT 기술은 최근 빌딩, 도시 관제 시스템뿐만 아니라 산업 현장의 설비와 인력관리 등 IoT 기술의 활용이 활발해지는 추세다. 본 논문에서는 IoT를 활용한 기술 개발을 통해 다양한 분야의 산업현장에 산재하고 있는 각종 위험인자를 포함하여 작업자의 근로 환경 정보를 영상 및 센싱 데이터를 이용하여 인지하고 작업자에게 현장 위해요인을 파악함으로써 합리적인 대책의 마련을 통해 작업자의 안전을 보장하고, 이와 더불어 업무지원 정보를 실시간으로 제공하여 인명피해의 감소, 작업능률과 생산성 향상을 야기할 수 있는 스마트 안전제어 및 원격 업무 지원을 위한 지능형 산업안전 관리 시스템을 개발하였다.

• Journal of Engineering Education Research
• /
• v.21 no.3
• /
• pp.66-75
• /
• 2018

This paper is the second part of the January 2018 issue of the Korean Society for Engineering Education, The "Equilibrium and Unbalance Analysis of Taegeuk Pattern DNA Matrix Codes," and is an extension of the paper published in the IoT Section of the 2017 Summer Conference in Jeju. In this paper, we have reviewed the history of what is life, and 5G Mobile communication: with IoT followed by recent research on influenza RNA gene mutation and DNA mutation variants, and the insights of Watson and Crick. Inspired by a single Franklin DNA X-ray diffraction photograph, they received the Nobel Prize for the Nature publication of DNA that has three patterns and regular repeatability. Professor MoonHo Lee has solved the three patterns in Diagonal, Left to Right, and Vertical matrices in a 2x2 matrix[CU; AG] and A = T = U = 30% C = G = 20%. We also proposed DNA Watch. This is the Healthcare IoT, which is seen by the DNA Watch on the wrist, the type of Tai Chi pattern of the body, and is immediately connected to the smartphone and delivered to the doctor.

• Journal of Internet of Things and Convergence
• /
• v.6 no.2
• /
• pp.25-29
• /
• 2020

The elliptic curve crypto-algorithm is widely used in authentication for IoT environment, since it has small key size and low communication overhead compare to the RSA public key algorithm. If the scalar multiplication, a core operation of the elliptic curve crypto-algorithm, is not implemented securely, attackers can find the secret key to use simple power analysis or differential power analysis. In this paper, an elliptic curve scalar multiplication algorithm using a randomized scalar and an elliptic curve point blinding is suggested. It is resistant to power analysis but does not significantly reduce efficiency. Given a random r and an elliptic curve random point R, the elliptic scalar multiplication kP = u(P+R)-vR is calculated by using the regular variant Shamir's double ladder algorithm, where l+20-bit u≡rn+k(modn) and v≡rn-k(modn) using 2^lP=∓cP for the case of the order n=2^l±c.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.9
• /
• pp.37-47
• /
• 2021

In this paper, we propose a new deep learning model for tongue segmentation with improved accuracy compared to the existing model by diversifying the receptive field in the U-net. Methods such as parallel convolution, dilated convolution, and constant channel increase were used to diversify the receptive field. For the proposed deep learning model, a tongue region segmentation experiment was performed on two test datasets. The training image and the test image are similar in TestSet1 and they are not in TestSet2. Experimental results show that segmentation performance improved as the receptive field was diversified. The mIoU value of the proposed method was 98.14% for TestSet1 and 91.90% for TestSet2 which was higher than the result of existing models such as U-net, DeepTongue, and TongueNet.

• Journal of the Korea Convergence Society
• /
• v.7 no.4
• /
• pp.19-24
• /
• 2016

To share the informations, we have been used Internet by the unique information transfer method. Many devices are showed and have been used in our various fields because of ICT(Information communication technology) development. The various devices are connected by Internet to send or receive the informations. We call this situation to The Internet of Things(IoT). It use the Internet for the interconnection of the various things. Studies on IoT progress extensively on today. Therefore, we studied a healthcare field related with Internet of things. In this paper, we described many actual cases such as smart bottle of medicine and smart pill in healthcare domains based on IoT.

• Science of Emotion and Sensibility
• /
• v.22 no.1
• /
• pp.23-34
• /
• 2019

The U-Healthcare era has evolved with the development of the Internet of things (IoT) in the early stages of being connected as a society. Already, many changes such as increased well-being and the extension of human life are becoming evident across cultures. Korea entered the growing group of aging societies in 2017, and its silver industry is expected to grow rapidly by adopting the IoT of a super-connected society. In particular, the senior shift phenomenon has resulted in increased interest in the promotion of the health and well-being of the emergent silver generation which, unlike the existing silver generation, is highly active and wields great economic power. This study conducted in-depth interviews to investigate the characteristics of the new silver generation, and to develop the design for a wearable serious game that intends to boost the interest of the elderly in exercise and fitness activities according to their personalized physical training regimes as prescribed by the U-Hospital service. The usage scenario of this wearable serious game for the 'U-silver generation' is derived from social necessity. Medical professionals can utilize this technology to conduct health examinations and to monitor the rehabilitation of senior patients. The elderly can also use this tool to request checkups or to interface with their healthcare providers. The wearable serious game is further aimed at mitigating concerns about the deterioration of the physical functions of the silver generation by applying personalized exercise prescriptions. The present investigation revealed that it is necessary to merge the on / off line community activities to meet the silver generation's daily needs for connection and friendship. Further, the sustainability of the serious game must be enhanced through the inculcation of a sense of accomplishment as a player rises through the levels of the game. The proposed wearable serious game is designed specifically for the silver generation that is inexperienced in using digital devices: simple game rules are applied to a familiar interface grounded on the gourmet travels preferred by the target players to increase usability.

• Information and Communications Magazine
• /
• v.32 no.4
• /
• pp.16-22
• /
• 2015

유비쿼터스 정보 기술을 기반으로 다양한 업무 지원 기기간 네트워크를 형성하여 업무 처리를 지원하는 u-Office와, 지능화된 다양한 사물에 인터넷 연결성을 제공하여 지능적 서비스를 제공하는 Internet of Things(IoT)는 기하급수적으로 증가하는 다수의 객체간 통신을 수행하므로 네트워킹 자원의 최적화 기술이 필수로 뒷받침되어야 한다. 데이터 패킷의 오버헤드를 줄이고 통신 자원을 절약하기 위한 기술 연구는 다방면으로 진행되고 있으며, 특히 패킷의 오버헤드를 줄이고 통신 자원을 절약하는 패킷 헤더 압축 기술은 다양한 국제 표준이 개발되어 모바일 와이맥스(Mobile WiMAX), Long Term Evolution(LTE), IPv6 over Low power Wireless Personal Area Network(6LoWPAN) 등의 네트워크에 활용되고 있다. 본 고에서는 모바일 와이맥스, LTE, 6LoWPAN에서 각각 사용되는 패킷 헤더 압축 기술인 Payload Header Suppression(PHS), Robust Header Compression(ROHC), Header Compression(HC), Improved Header Compression(IPHC)의 개발 동향을 제시하고 각 기술의 특징을 비교 분석한다.