• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.185-192
• /
• 2021

Various methods have been proposed to determine the defect detection of welding bead, and recently sensor data and image data inspection have been steadily announced. There are advantages that sensor data inspection is highly accurate, and two-dimensional-based image data inspection is able to determine the position of the welding bead. However, when analyzing only with sensor data, it is difficult to determine whether the welding has been performed at the correct position. On the other hand, the image data inspection does not have high accuracy due to noise and measurement errors. In this paper, we propose a method that can complement the shortcomings of each inspection method and increase its advantages to improve accuracy and speed up inspection by fusing sensor data inspection which are average current, average volt, and mixed gas data, and image data inspection methods and is implemented as software. In addition, it is intended to allow users to conveniently and intuitively analyze and grasp the results by performing analysis using a graphical user interface(GUI) and checking the data and inspection results used for the inspection. Sensor inspection is performed using the characteristics of each sensor data, and image data is inspected by applying a morphology geodesic active contour algorithm. The experimental results showed 98% accuracy, and when performing the inspection on the four image data, and sensor data the inspection time was about 1.9 seconds, indicating the performance of software that can be used as a real-time inspector in the welding process.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.9
• /
• pp.595-602
• /
• 2013

WiFi fingerprinting is well known as an effective localization technique used for indoor environments. However, this technique requires a large amount of pre-built fingerprint maps over the entire space. Moreover, due to environmental changes, these maps have to be newly built or updated periodically by experts. As a way to avoid this problem, crowd-sourced fingerprint mapping attracts many interests from researchers. This approach supports many volunteer users to share their WiFi fingerprints collected at a specific environment. Therefore, crowd-sourced fingerprinting can automatically update fingerprint maps up-to-date. In most previous systems, however, individual users were asked to enter their positions manually to build their local fingerprint maps. Moreover, the systems do not have any principled mechanism to keep fingerprint maps clean by detecting and filtering out erroneous fingerprints collected from multiple users. In this paper, we present the design of a crowd-sourced fingerprint mapping and localization(CMAL) system. The proposed system can not only automatically build and/or update WiFi fingerprint maps from fingerprint collections provided by multiple smartphone users, but also simultaneously track their positions using the up-to-date maps. The CMAL system consists of multiple clients to work on individual smartphones to collect fingerprints and a central server to maintain a database of fingerprint maps. Each client contains a particle filter-based WiFi SLAM engine, tracking the smartphone user's position and building each local fingerprint map. The server of our system adopts a Gaussian interpolation-based error filtering algorithm to maintain the integrity of fingerprint maps. Through various experiments, we show the high performance of our system.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.32 no.4
• /
• pp.340-355
• /
• 1996

In order to suggest the useful information of fishing ground, a data base system on 32bit personal computer was constructed and handled by using the catch data of Korean tuna long-line, catch by species, fishing time and place, fish price and etc. mainly from 1975 to 1992. The results obtained are as follows ; In the fishing ground displaying catch rate, the catch rate has reduced as time passed, and this penomenon was more evident in Indian. And yellowfin have high catch tate in the Western Pacific of low latitute region, bigeye tuna have in the Eastern. The region of high catch rate of bigeye tuna was moved from the Indian and the Atlantic to the Pacific. The patterns of catch numbers of yellowfin and bigeye tuna appeared nearly same that, having nothing to do with catch numbers in all oceans. The region of least catch was the Northwestern Pacific, the regions of most catch were the Western Indian and the Pacific of low latitute. As to simulation of fishing ground estimation, there were economical grounds in the Western Pacific of low latitute region, the Eastern Pacific of this, the Western Indian, the Eastern Indian, and the Atlantic, in order.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.3
• /
• pp.461-467
• /
• 2019

Advances in location-aware and IoT (Internet of Things) technology increase the rapid generation of massive movement data. Knowledge discovery from massive movement data helps us to understand the urban flow and traffic management. This paper proposes a method to analyze frequent origin-destination sequence patterns from irregular spatiotemporal taxi pick-up locations. The proposed method starts by conducting cluster analysis and then run a frequent sequence pattern analysis based on identified clusters as a base unit. The experimental data is Seoul taxi trajectory data between 7 a.m. and 9 a.m. during one week. The experimental results present that significant frequent sequence patterns occur within Gangnam. The significant frequent sequence patterns of different regions are identified between Gangnam and Seoul City Hall area. Further, this study uses administrative boundaries as a base unit. The results based on administrative boundaries fails to detect the frequent sequence patterns between different regions. The proposed method can be applied to decrease not only taxis' empty-loaded rate, but also improve urban flow management.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.10
• /
• pp.493-498
• /
• 2017

Capsule endoscopy is specialized for the observation of small intestine that is difficult to access by general endoscopy. The diagnostic procedure through capsule endoscopy consists of three stages: examination of indicant, endoscopy, and diagnosis. At this time, key information needed for diagnosis includes indicant, lesions, and suspected disease information. In this paper, these information are defined as semantic features and the extracting process is defined as semantic-based analysis. It is performed in whole capsule endoscopy. First, several symptoms of patient are checked before capsule endoscopy to get some information on suspected disease. Next, capsule endoscopy is performed by checking the suspected diseases. Finally, diagnosis is concluded by using supporting information. At this time, some association are used to conclude diagnosis. For example, there are the disease association between the symptom and the disease to identify the expected disease, and the anatomical association between the location of the lesion and supporting information. However, existing knowledge models such as MST and CEST only lists the simple term related to endoscopy and cannot consider such semantic associations. Therefore, in this paper, we propose association-based knowledge model for supporting diagnosis of capsule endoscopy. The proposed model is divided into two; a disease model and anatomical model of small intestine, interesting area(organs) of capsule endoscopy. It can effectively support diagnosis by providing key information for capsule endoscopy.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.4
• /
• pp.413-422
• /
• 2012

After development of GPS in the 1960's, the United States released SA(Selective Availability) in 2000 and then the GPS has become commercialized to the present. The result of repeatedly developed GPS observation, the GPS real-time observation methods is RTK which basically always needs two base stations and has a fault of the accuracy decreasing as the distance between a mobile station and a receiver is increasing. Because of these weakness, VRS method has come out. VRS(Virtual Reference Station) generates the imaginary point near mobile station from several observatory datum of GPS, sets the accurate location of mobile station, thus shows high reliability and mobility. Now, the cadastral datum point is used with azimuth, repetition, and graphical traversing method for traverse network. The result of measurement indicates many problems because of different accomplishment interval given point, restrictions on the length of the conductor, many errors on the observations. So, this study did comparative analysis of the cadastral datum points through VRS method by Continuously Operating Reference Station. Through the above comparative analysis, The comparative result between surveyed result with repetition method through total station observed Cadastral Control Points and surveyed result with VRS-RTK has shown that average error of x-axis is -0.08m, average error of y-axis, +0.07m and average distance error is +0.11m.

• Korean Journal of Plant Resources
• /
• v.33 no.1
• /
• pp.1-14
• /
• 2020

This study was conducted to analyze community structure and species composition of Pinus densiflora for. erecta Stand in Mt. Cheonchuk (653 m). Field survey was carried out from June to September in 2013. 74 plots (20×20 m) were set up, 5 herb layer plots (3×3 m) were constructed for each plot, and there, Diameter at Breast Heigh t(DBH), height, environmental factor, annual growth were measured. Vascular plants were surveyed as following; 66 family, 165 genus, 211 species, 2 sub species, 29 variety, 6 form 248 taxa. Results of cluster analysis for P. densiflora for. erecta forest, 3 communities were divided into; Quercus mongolica (P-1), Quercus variabilis (P-2) and Quercus aliena-Stephanandra incisa (P-3). There were significant environmental factors that organic layer, annual growth, CEC, total total nitrogen, organic matter and pH for each community. As a result of DCA, P-1 and P-2 were distributed large range of environmental factors but relatively limited in P-3. Distributions of herb layer were affected by sand, cation exchange capacity, silt and total nitrogen. Results of MRPP test for herb layer communities, it was significantly analyzed (A=0.003, P<0.008). Species diversity index was highly recorded in P-3 and influenced by cation exchange capacity, total nitrogen, annual growth in consequence of NMS analysis.

• Proceedings of the KWS Conference
• /
• 2010.05a
• /
• pp.49-49
• /
• 2010

용접은 산업계의 기계 조립 및 접합을 위한 공정의 주요한 작업으로 조선, 중공업, 건설 등 산업현장에서 사람에 의한 수동적인 작업으로 대부분 수행된다. 이러한 용접 작업을 수행하는 용접 기술자는 산업 현장 훈련원과 직업 교육 학교에서 양성되지만 용접 훈련 과정은 실습 초보자에게 위험하고, 장시간 교육하기에 어려울 뿐 아니라 재료 낭비, 의사 소통의 한계, 즉석 결과 평가의 한계, 공간부족 등 다양한 문제가 있다. 그러므로, 안전하고 반복적인 실습 환경 제공하고 장시간 및 다수 교육참여 지원 등이 가능한 시스템을 구축하여 숙련된 우수 인력 조기 확보와 훈련 비용을 절감할 필요가 있다. 본 논문에서는 실제와 동일한 상호작용을 제공할 뿐만 아니라 고품질로 훈련 환경을 가시화하여 용접 상황을 동일하게 모사하는 가상 현실 기반 용접 훈련 시뮬레이터를 제시한다. 이 시스템은 용접의 형상과 환경의 고품질 가시화, 경험 DB를 통한 용접의 비드 형상 데이터 획득, 용접 토치를 이용하는 사용자 상호작용, 용접 훈련 결과 평가 및 최적 작업 가이드, 용접 콘텐츠 저작, 다양한 용접 훈련을 가시화하는 하드웨어 플랫폼으로 구성된다. 고품질 가상 용접 가시화는 경험 DB 기반 비드 형상 데이터와 신경회로망을 이용한 비드 형상 예측을 통해 실시간 비드 표현이 이루어지며 쉐이더 기반 고품질 모재 및 비드 표현, 아크 불꽃 효과 표현을 포함한다. 사용자 상호작용은 현장 작업 도구와 일치된 토치 인터페이스와 위치추적을 이용하여 토치의 작업각, 진행각, 속도, 거리 등을 반영할 수 있으며 진동과 소리 등 용접 훈련의 사실적 상호작용도 재현하였다. 용접 훈련 평가 및 최적 작업 가이드는 훈련자의 용접속도, 거리, 각도 등의 사용자 작업 결과를 그래픽으로 표현하고, 애니메이션을 통한 훈련 자세를 추후 분석할 수 있도록 하였고, 가상토치, 기준선, 수치계기 등을 이용한 최적 작업 훈련 가이드 제시하였다. 훈련 콘텐츠 저작은 메뉴UI 기반으로 용접의 전류, 전압 등의 조건과 상황을 선택하도록 제시하였고, 하드웨어 플랫폼은 워크벤치형 입체 디스플레이 방식으로 용접 환경을 가시화하였고, 위, 정면, 아래보기 등 다양한 용접 자세 변경을 지원 할 수 있도록 구축하였다. 이러한 가상현실 기반 훈련 시뮬레이터는 아크열 발생에 따른 장시간 훈련의 어려움을 극복할 수 있고, 다양한 실습 환경을 바꾸어 가며 반복적인 훈련이 가능하고, 실 재료를 사용하지 않아 재료의 낭비를 줄일 수 있는 환경 친화적인 안전하고 효율적인 훈련 실습 환경을 제공할 수 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.489-492
• /
• 2006

Red tide occurred sporadically in early 90s. But It is happening extensively by global warming. So, Airline observation, Red tide buoy development, and Red tide alarm system research is progressing for monitor ring. However, study to early forecast red tide and red tide alarm system did not exist hard. This paper proposed development that design and implementation red tide database of using wireless sensor network. There are GPS, Water Temperature sensor, Oxygen sensor, and Turbidity sensor in each node. And data is stored to red tide database through Ad-hoc network. This data is integrated and analyzed. So, forecast red tide. And red tide database has red tide data that happen at past. This is utilized to comparative analysis data for red tide estimate. Main screen displays position of node and measured value in electron map. Much studies must be backed for this a study. But I think that contribute to analyze red tide data by red tide database construction.

• Proceedings of the KSR Conference
• /
• 2008.11b
• /
• pp.1069-1077
• /
• 2008

The railway abrasion measurement system have to satisfy two conditions to increase the measurement accuracy as follows. The laser region which is projected on the rail have to be extracted without the geometrical distortion. The mapping of the acquired laser region data on the rail profile have to be processed exactly. But, the conventional railway abrasion measurement system is deeply effected by the foreign substance( dust, rainwater, and so on ) on the railway or the sensitive response characteristic of the laser to the external measurement circumstance, and then the measurement errors arise from above factors. When the laser region is projected on the rail extracts from the acquired image, the interference of the light with the same frequency as the laser system occurs the serious problems. In the process of the mapping between the railway profile and the extracted laser region, the measurement accuracy is very highly effected by the geometrical distortion and the abnormal variation. In this Paper, we propose the novel method to increase the accuracy of the railway abrasion measurement dramatically. we designed and manufactured the high precision and fast image processing board with DSP Core and FPGA to measure the railway abrasion. The image processing board has the capability that the image of 1024X1280 from camera can be processed with the speed of 480 frame/sec. And, we apply the image processing algorithm base on the wavelet to extract the laser region is projected on the rail exactly. Finally, we developed high precision railway abrasion measurement system with the error range less than +/-0.5mm by which 2D image data is covered 3D data and mapped on the rail profile using the camera model and the perspective transform.