• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.452-455
• /
• 2017

As the number of high - rise buildings has increased recently, this research society has become interested in the method of cleaning the external wall required. When cleaning these areas, an accident occurs every year when a person uses dangerous manual labor using a gondola. The main goal of this work is to enable people to manipulate the robot with simple operation without dangerous manual operation when working in a vertical structure. As the altitude increases, the concept of the mechanism attaching to the vertical structure while enduring the external resistance in the increase of the wind strength is applied, and the additional attachment device is attached to the end of the leg based on the existing four- According to the control, the development result of the robot having the function of detachable to the glass is presented.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.2
• /
• pp.409-415
• /
• 2017

This paper proposes a road segmentation algorithm using a watershed. The proposed algorithm is a segmentation algorithm using an automatic marked watershed that automatically creates a road marker and a background marker using information about vehicles and lanes on road and it can solve problems of a watershed-based segmentation such as overmany regions or handworks for markers. The road marker has property for pure road areas in which lanes are included but vehicles are excluded and the background marker has property for the areas left in which vehicles and background are included. Results of segmentation applied to real road images show that the proposed algorithm can automatically creates appropriate markers and it can properly segments the required road area that include the lane with a vehicle and its both side lanes in various environments, and it is equal to the conventional algorithm using markers created by handwork in performance.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.197-199
• /
• 2000

연속음성인식을 수행하기 위해서는 발음사전과 언어모델이 필요하다. 이 둘 사이에는 디코딩 단위가 일치하여야 하므로 발음사전 구축시 디코딩 단위로 표제어 단위를 선정하며 표제어 사이의 음운변화 현상을 반영한 발음사전을 구축하여야 한다. 한국어에 부합하는 음운변화현상을 분석하여 학습용 자동 발음열을 생성하고, 이를 통하여 발음사전을 구축한다. 전처리 단계로 기호, 단위, 숫자 등 전처리 과정 및 형태소 분석 과정을 수행하며, 디코딩 단위인 의사 형태소 단위를 생성하기 위해 규칙을 이용한 태깅 과정을 거친다. 이를 통해 나온 결과를 발음열 생성기 입력으로 하며, 결과는 학습용 발음열 또는 발음사전 구성을 위한 형태로 출력한다. 표제어간 음운변화 현상이 반영된 상태의 표제어 단위이므로 실제 음운변화가 반영되지 않은 상태의 표제어와는 그 형태가 상이하다. 이는 연속 발음시 생기는 현상으로 실제 인식에는 이 음운변화 현상이 반영된 사전이 필요하게 된다. 생성된 발음사전의 효용성을 확인하기 위해 다음과 같은 실험을 통해 성능을 평가하였다. 음향학습을 위하여 PBS(Phonetically Balanced Sentence) 낭독체 17200문장을 녹음하고 그 전사파일을 사용하여 학습을 수행하였고, 발음사전의 평가를 위하여 이 중 각각 3100문장을 사용하여 다음과 같은 실험을 수행하였다. 형태소 태그정보를 이용하여 표제어간 음운변화 현상을 반영한 최적의 발음사전과 다중 발음사전, 언어학적 기준에 의한 수작업으로 생성한 표준 발음사전, 그리고 표제어간의 음운변화 현상을 고려하지 않고 독립된 단어로 생성한 발음사전과의 비교 실험을 수행하였다. 실험결과 표제어간 음운변화 현상을 반영하지 않은 경우 단어 인식률이 43.21%인 반면 표제어간 음운변화 현상을 반영한 1-Best 사전의 경우 48.99%, Multi 사전의 경우 50.19%로 인식률이 5~6%정도 향상되었음을 볼 수 있었고, 수작업에 의한 표준발음사전의 단어 인식률 45.90% 보다도 약 3~4% 좋은 성능을 보였다.

• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.689-698
• /
• 2015

Currently, the itemized box loading operation at the Agriculture Products Processing Center which distributes agricultural products for the region is carried out manually. The process of loading agricultural products requires great manpower and had been resolved through the part-time employment of the residents of farm villages. However, recently it has become quite difficult to secure manpower as the aging within the rural community has been intensified. Hence, the necessity for countermeasures such as facility automation or use of robots have become necessary. This study suggests an automatic robot system for the itemized loading of apple boxes at the Agriculture Products Processing Center. The suggested method is to design and develop equipment such as a conveyer, robot, and bar code reader. In addition, a sorting plan, operational control, generation of control information, and software module that could monitor the inside of the Agriculture Products Processing Center is needed. After test-operating and evaluating the developed system, the existing manual work would be replaced with the automated robot system in order to enhance work efficiency and resolve safety issues.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.7
• /
• pp.1713-1723
• /
• 1999

Optical character recognition(OCR) might be the most plausible method in building database out of printed matters. This paper describes the points to be considered when one selects an OCR system in order to build database. Based on the considerations, we evaluated four commercial OCR systems, and chose one which shows the best recognition rate to build OCT-text database. The subject text, the KT-test collection, is a set of abstracts from proceedings of different printing quality, fonts, and formats. KT-test collection is also provided with typed text database. Recognition rate was calculated by comparing the recognition result with the typed text. No preprocessing such as learning and slant correction was applied to the recognition process in order to simulate a practical environment. The result shows 90.5% of character recognition rate over 970 abstracts. This recognition rate is still insufficient for practical use. The errors in OCR texts are different from those of manually typed texts. In this paper, we classify the errors in OCR texts for the further research.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.17 no.3
• /
• pp.245-256
• /
• 1999

Watershed delineation is the process of tracing surface water flow across a landscape to identify hydrologic basins. Manual and automated techniques have been developed to trace surface waters and locate watersheds on a variety of landscape. Low topographic relief landscapes, however, are particulary problematic for automated watershed delineation techniques. Therefore, this study aims at focusing the watershed delineation algorithm of low topographic relief landscape. By using topogrid, which manages enforcement drainage algorithms and incorporates topographic data into stream data, this study delineates watershed boundary in low relief landscape. Based on this research, in a low topographic relief landscape, enforcement drainage algorithms using topogrid generate better source grid to delineate watershed boundary than using only TIN. Also, comparing automated watershed delineation which incorporates contours and elevations into stream data with manual watershed delineation, we can know that the incorporation of both elevation and stream data generate more effective results.

• Journal of Korea Society of Industrial Information Systems
• /
• v.13 no.2
• /
• pp.88-99
• /
• 2008

In this article, DHRMIS being developed with ERP at resent realizes the personnel affairs recording procedures that all the Ministry of National Defense, the army, the navy and the air force can use on personnel affairs in order to solve the problem of the personnel recording procedures of the MND and each of the troops such as using manual affairs together with computation affairs, absence of electronic recording managements in some of the institutions MND, insufficiency of real time recording, and excess of centralized managements to present and record the personnel data. The personnel recording procedures consist of standardized personnel recording procedures and definition of the personnel recording procedures. This personnel recording procedures propose 3 improvements such as an atomic electronic recording management, a real time recording affair, effective centralization and distribution of verification authority so as to guarantee authentication, integrity, reliance, and accessibility of personnel data. When proposed personnel recording procedures are realized through developing DHRMS, these can bring on unification of 10 recording management systems, reduction of recording management time from 45 days to 3 days, distribution of verification authority MND, each of the troops, each of the institutes, an individual, and reduction of some of the manual documents.

• The Journal of the Korea Contents Association
• /
• v.17 no.5
• /
• pp.221-233
• /
• 2017

To survive in the ever-changing environment, organizations need to improve or innovate their business processes. As a result, to attain this objective, BPM (Business Process Management) concept is widely adopted in modern enterprises. BPM life cycle consists of diagnosis, design, implementation and enactment. Conventionally, diagnosis of business process within the BPM life cycle is usually conducted by manual methods such as interviews, questionnaires and direct observations of process. And (re)designing business processes is also usually done manually under supervision of business experts from scratch. It is time-consuming and error-prone tasks. The objective of this research is to integrate the diagnosis and (re)design phase of BPM life cycle by sharing automatically generated process model and basic statistics in the diagnosis phase based on the process mining method. Eventually, this approach will lead to automate the tasks of diagnosis and design of business process. To implement and to show the usefulness of the proposed framework, two case studies were conducted in this research.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.9
• /
• pp.4192-4198
• /
• 2011

Terrestrial LiDAR is a high precision positioning technique to monitor the behavior and change of structures and natural slopes, but it has depended on subjective hand intensive tasks for the classification(surface and vegetation or structure and vegetation) of positioning data. Thus it has a couple of problems including lower reliability of data classification and longer operation hours due to the surface characteristics of various geographical and natural features. In order to solve those problems, the investigator developed a technique of using the NDVI, which is a major index to monitor the changes on the surface(including vegetation), to categorize land covers, combining the results with the terrestrial LiDAR data, and classifying the results according to items. The application results of the developed technique show that the accuracy of convergence was 94% even though there was a problem with partial misclassification of 0.003% along the boundaries between items. The technique took less time for data processing than the old hand intensive task and improved in accuracy, thus increasing its utilization across a range of fields.

• The Journal of the Korea Contents Association
• /
• v.6 no.6
• /
• pp.1-8
• /
• 2006

This paper presents an intelligent pattern matching method that can be used to acquire the reliable calibration data for automatic PCB pattern inspection. The inaccurate calibration data is often acquired by geometric pattern variations and selecting an inappropriate model manual. It makes low the confidence of inspection and also the inspection processing time has been delayed. In this paper, the geometric features of PCB patterns are utilized to calculate the accurate calibration data. An appropriate model is selected automatically based on the geometric features, and then the calibration data to be invariant to the geometric variations(translation, rotation, scaling) is calculated. The method can save the inspection time unnecessary by eliminating the need for manual model selection. As the result, it makes a fast, accurate and reliable inspection of PCB patterns.