Earticle

Currently, Korea government is pushing to reduce the energy of various building to meet GHG(Greenhouse Gas) reduction target. There are various methods to save energy in order to fulfill this reduction obligation. One of the ways is detection of spaces using IoT environment sensors to save energy on buildings based on monitoring. Some spaces waste the energy to meet thermal comfort level even when there is no people. In this paper, we show the practical results of our occupancy detection algorithm using IoT environment sensors to find the wasted space on the building.

최근 과학기술통신부에서 발표한 '디지털 뉴딜'의 10대 대표 과제 중 하나인 '데이터 댐' 사업에 높은 비용이 투자되고 있다. 이에 따른 AI 학습용 데이터 구축 관련 국내 정책 및 사업을 분석해 보면 데이터 레이블링 자동화 부분에 부족함이 존재한다. 데이터 레 이블링을 수작업으로만 진행하기에는 노동력이 많이 필요하고 시간도 오래 걸린다. 그뿐만 아니라, 사람마다 다른 레이블링 방식은 데이터의 일관성을 떨어뜨려 모델 학습에 부정적인 영향을 준다. 따라서 본 연구는 데이터 구축 시간 단축과 노동력 절감, 일관성 있는 레이블링으로 빠르게 양질의 데이터 추출이 가능하다. 또한, 전이학습을 통해 AI 모델의 정확도를 향상하여 수작업 빈도를 줄 여 단시간에 대량의 데이터를 추출하는 것에 도움이 되며, 새로운 객체의 인식 또한 가능하여 다양한 객체 탐지에 도움을 줄 수 있 다. 이미지 레이블링은 KEM 형식으로 생성되기 때문에 생성된 데이터는 교육 분야에도 유의미한 데이터로 사용될 수 있다.

High costs are being invested in the "Data Dam" project, one of the top 10 representative tasks of the "Digital New Deal" recently announced by the Ministry of Science and ICT. Accordingly, as a result of analyzing domestic policies and projects related to data construction for AI learning, there is a lack in data labeling automation. Data labeling requires a lot of labor and takes a long time to proceed manually. In addition, different labeling methods from person to person have a negative impact on model learning due to poor data consistency. Therefore, in this study, it is possible to quickly extract high-quality data by shortening data construction time, reducing labor, and consistent labeling. In addition, it helps to extract data in a short time by reducing the frequency of manual work by improving the accuracy of the AI model through transfer learning, and it can help detect various objects by enabling recognition of new objects. Since image labeling is generated in the KEM format, the generated data may also be used as meaningful data in the field of education.

최근 제주 지역은 2024년 기준 전체 혼인의 13.3%를 차지할 만큼 다문화 인구 비중이 급격히 확대되고 있으며, 이로 인해 가정 내 이중언어 사용과 지역사회의 제주 방언이 혼재하는 독특하고 복합적인 언어 환경이 형성되고 있다. 제주 다문화 아동들은 이러한 다층적인 환경에서 성장하고 있으나, 기존의 표준화된 검사 도구는 단일 언어 사용자를 기준으로 설계되어 이들의 언어적 잠재력과 차이를 정확히 변별하는 데 한계가 있다. 이에 가장 타당한 평가 방법론으로 간주되는 자발화 표본 분석(LSA)은 전사 및 분석에 소요되 는 막대한 시간적 부담과 참조 규준의 부재로 인해 임상 현장에서의 활용이 제한적인 실정이다. 이에 본 연구는 다문화 아동의 언어 능력을 타당하고 효율적으로 평가하기 위한 ‘제주 형 CHILDES CHAT 표기 규약’ 시안과 이를 구현할 기술적 파이프라인을 제안한다. 구체적으로 본 규약은 데이터의 국제적 상호운용성을 확보하기 위해 ISO 3166-1 및 ISO 639-3 국제 표준 코드를 준용하여 연구 대상을 베트남(MC-VN), 중국(MC-CN) 등으로 체계화하 였다. 특히 중국계 가정의 경우 한족(MC-CN-H)과 한국계 중국인(MC-CN-K)으로 세분 화하여 연변 방언 등의 언어적 변인을 정밀하게 통제할 수 있도록 설계하였다. 또한, 이중 언어 간섭을 분석하기 위한 코드 스위칭 마커(@s)와 지역 방언 사용 비중을 측정하기 위한 제주어 마커(@d), 그리고 연변 방언 구분을 위한 확장 마커(@d:yb)를 도입하여 다층 적인 언어 사용 양상을 정량화할 수 있는 기틀을 마련하였다. 더불어 본 연구는 이러한 정밀한 전사 과정이 연구 현장에서 실질적으로 작동할 수 있도록 ‘AI 기반 자동화 파이프라인’을 함께 구축하였다. 이는 최신 AI 음성 인식 기술과 고성능 외부 형태소 분석기를 연동하여 전사 시간을 획기적으로 단축하고 분석의 정확도 를 높이는 하이브리드 분석 아키텍처이다. 본 연구가 제안하는 표준화된 규약과 자동화된 방법론은 향후 구축될 ‘제주 다문화 아동 언어 말뭉치’의 견고한 초석이 될 것이며, 이를 통해 대상 아동의 특성에 부합하는 맞춤형 임상 진단 및 교육적 지원을 실현하는 데 기여 하고자 한다.

With multicultural marriages accounting for 13.3% of the total in 2024, Jeju is experiencing a rapid demographic shift, creating a complex linguistic environment where bilingualism coexists with the local dialect. Existing assessment tools, designed for monolinguals, fail to accurately capture the linguistic potential of multicultural children in this setting. Even Language Sample Analysis, considered the most valid method, is limited in clinical use by significant time constraints and a lack of norms. In response, this study proposes the "Jeju-specific CHILDES CHAT Transcription Conventions" and an automated pipeline for efficient assessment. Adhering to ISO 3166-1 and ISO 639-3 standards for interoperability, the conventions categorize subjects by background (e.g., MC-VN, MC-CN), notably subdividing Chinese families into Han (MC-CN-H) and Korean-Chinese (MCCN- K) to control for variables like the Yanbian dialect. Specific markers are introduced to quantify language patterns: @s for bilingual interference, @d for Jeju dialect density, and @d:yb to distinguish the Yanbian dialect. To ensure practical application, the study also presents an "AI-based Automation Pipeline." This hybrid architecture integrates AI speech recognition with external morphological analyzers, drastically reducing transcription time while enhancing accuracy. These standardized conventions and automated methodologies will serve as a foundation for the future "Jeju Multicultural Children's Language Corpus," facilitating tailored clinical diagnosis and educational support.

The human centric lighting (HCL) control is a major focus point of the smart lighting system design to provide energy efficient and people mood rhythmic motivation lighting in smart buildings. This paper proposes the HCL control using indoor surveillance camera to improve the human motivation and well-beings in the indoor environments like residential and industrial buildings. In this proposed approach, the indoor surveillance camera video streams are used to predict the day lights and occupancy, occupancy specific emotional features predictions using the advanced computer vision techniques, and this human centric features are transmitted to the smart building light management system. The smart building light management system connected with internet of things (IoT) featured lighting devices and controls the light illumination of the objective human specific lighting devices. The proposed concept experimental model implemented using RGB LED lighting devices connected with IoT features open-source controller in the network along with networked video surveillance solution. The experiment results are verified with custom made automatic lighting control demon application integrated with OpenCV framework based computer vision methods to predict the human centric features and based on the estimated features the lighting illumination level and colors are controlled automatically. The experiment results received from the demon system are analyzed and used for the real-time development of a lighting system control strategy.

KNXnet/IP protocol defines mechanisms for using KNX Home and Building Automation technology on top of the IP protocol using client/server architecture. Due the nature of KNXnet/IP protocol it presents some connectivity and security problems inherent to the protocol definition. In this work we propose some mechanisms for trying to fix these problems and finally we implement and test them using real KNX devices.

The building energy management systems its control and automation in buildings has significant role. These systems can play an important role in regular energy monitoring and management and therefore to save the possible energy and cost. The key point of the building automation market is focused upon better facilitation to the user in terms of comfort at reduced operation cost. Energy efficiency improvement will also contribute to environmental protection. Therefore there have been regulations and rating systems made that mandates the requirement of energy monitoring and control in a building. For example, the above mentioned building utilities and equipment control and automation plays an integral role in achieving the green building rating points from certifying authorities such as GRIHA and IGBC. The proposed system is to control the active systems such as lighting including artificial lighting (on/off & dimming control), air conditioners and safety features like fire alarm & gas alarm. In future the existing idea can be implemented for the whole building, i.e. various rooms or areas and then all of them can be integrated on a common platform for monitoring and control of different equipment.

The building energy management systems its control and automation in buildings has significant role. These systems can play an important role in regular energy monitoring and management and therefore to save the possible energy and cost. The key point of the building automation market is focused upon better facilitation to the user in terms of comfort at reduced operation cost. Energy efficiency improvement will also contribute to environmental protection. Therefore there have been regulations and rating systems made that mandates the requirement of energy monitoring and control in a building. For example, the above mentioned building utilities and equipment control and automation plays an integral role in achieving the green building rating points from certifying authorities such as GRIHA and IGBC. The proposed system is to control the active systems such as lighting including artificial lighting (on/off & dimming control), air conditioners and safety features like fire alarm & gas alarm. In future the existing idea can be implemented for the whole building, i.e. various rooms or areas and then all of them can be integrated on a common platform for monitoring and control of different equipment.

Home and building automation systems (HBAS) today are categorized by plethora of data formats, communication platforms and software services accessible via heterogeneous embedded devices. These heterogeneous embedded devices characterized a building into a data intensive entity, generating numerous data and complex control information, brings a much heavy load to the HBAS. Consequently, this leads to interoperation difficulties among those devices for information exchange and operation management. In this paper, a Service Oriented Architecture (SOA) based framework that leveraged by services is proposed to manage HBAS efficiently by reducing their workload during joint execution of tasks. The proposed framework consists of several layers dedicated for services interwoven by Event-Condition-Action (ECA) oriented decision mechanism for managing heterogeneous embedded devices. The proposed framework is viable and highly efficient to be deployed for HBAS with limited computing resources.

The purpose of this study is to develop the category, intelligent element technologies and detail items for application of intelligent school. The result of this study will be a basic data to improve the intelligent school. For this purpose, analyze domestic and foreign intelligent system and through interview, survey with professional group, developed item was evaluated and for verifying and reappraising the items the researcher visited and interviewed teachers of three elementary schools. Developed intelligent items were composed of twelve basic technologies and fifty-four detail technologies.

There are many economic and operational reasons to integrate fire alarm signaling system with other building automation system. Integration of this requires open network with for the standard communication protocol and careful design practices. The important point for this is also the development of intelligent control modules for replacing the conventional zone adapter in fire system. Therefore, this paper proposes an new conceptual design of the open distributed fire alarm signaling system for the integration of BAS, a new intelligent control modules with LonTalk protocol and a new control modules for power line communication. Newly proposed additions to LonWorks network make it very well suited for integrating fire systems with other building automation sub-systems and another BAS with low cost. Additionally, it is very important that best design practices, test procedures and building codes need to be modernized to accommodate integrated building systems.

This paper describes an automation of architectural design processes and the direction between this automation and robotic system in building construction using the concept of the object-oriented design in architecture. The research starts from the premise that many computer-aided design systems are mostly aimed at serving as drawing tools which are used only after a design formal solution has already been established by the architect. If the computer is well applied to the architectural design process, many positive things such as standardization of design detail, increasing productivity and collaboration, minimizing construction costs etc. can be done. To support an early design solution in the computer-based environment, the proposed automation was developed and tested in a real building project, using the object-oriented design concept such as product model. This design automation gives various design alternatives from the early design phase to the final stage of design details, according to musical harmony. This paper shows how architectural design process can be automated and how the data of the applied architectural design can be integrated into product model environment, in relation to robotic system in building construction.

System Integration is significant role in building, because it is that the more building system is integrated, the more they can save management costs and energy costs. Nowadays system protocols in building use BACnet, LonWorks, etc. to integrate effectively, but their interoperability is not perfect. Therefore, we propose system integration using Ethernet and TCP/IP for perfect harmony. When data network protocol, Ethernet and TCP/IP is used for control network, there are several problems. But there have been solved by many researches. therefore it is possible that building system integrate to use Ethernet and TCP/IP. Finally, we show that Ethernet and TCP/IP can be used for building automation system integration.

This paper summarizes economic considerations for building automation. While the market for energy management systems(EMS) has been exploding, there are many indications that the pendulum of hight expectations may have swung too far. This is a very important reason for energy management system in both existing and new buildings. Because facilities continue to expand in size and their operations grow more complex, the use of automated systems to control and monitor both programmable and nonprogrammed function becomes an economic necessity. The following is offered as engineerings of economic consideration for building automation systems. Not only get a complete return on investment but increase profits in a few years through savings incurred in labor, utilities fuel and extended equipment life.