• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제10권4호
• /
• pp.281-286
• /
• 2010

We propose a space and time based sensor fusion method and a robust landmark detecting algorithm based on sensor fusion for mobile robot navigation. To fully utilize the information from the sensors, first, this paper proposes a new sensor-fusion technique where the data sets for the previous moments are properly transformed and fused into the current data sets to enable an accurate measurement. Exploration of an unknown environment is an important task for the new generation of mobile robots. The mobile robots may navigate by means of a number of monitoring systems such as the sonar-sensing system or the visual-sensing system. The newly proposed, STSF (Space and Time Sensor Fusion) scheme is applied to landmark recognition for mobile robot navigation in an unstructured environment as well as structured environment, and the experimental results demonstrate the performances of the landmark recognition.

• 한국문헌정보학회지
• /
• 제54권2호
• /
• pp.419-434
• /
• 2020

본 연구의 목적은 랜드마크 이미지의 AI 학습용 데이터 구축을 위한 메타데이터 표준 설계 방안을 제시하기 위함이다. 이를 위해, 이미지 검색시스템의 종류와 각각의 색인 방식에 관한 최신 기술 현황을 포괄적으로 조사하여 분석하고, AI 머신러닝을 적용한 랜드마크 인식에 필수적인 학습용 공개 데이터셋과 이미지 객체 인식에 관한 기계학습 도구를 조사하였다. 이를 통해, 랜드마크 이미지 AI 학습용 데이터에 최적화된 메타데이터 요소를 선정하고 각각의 요소에 대한 입력 데이터를 정의하였다. 결론 및 제언에서는 랜드마크 인식을 활용한 추천시스템을 포함한 응용서비스 개발 방안을 논의하였다.

• 응용통계연구
• /
• 제31권5호
• /
• pp.539-554
• /
• 2018

최근 등장하는 다양한 사물인터넷 기기 혹은 상황인식 기반의 인공지능에서는 사용자와 기기의 상호작용이 중요시 된다. 특히 인간을 대상으로 상황에 맞는 대응을 하기 위해서는 인간의 표정을 실시간으로 인식하여 빠르고 정확한 판단을 내리는 것이 필요하다. 따라서, 보다 빠르고 정확하게 표정을 인식하는 시스템을 구축하기 위해 얼굴 이미지 분석에 대한 많은 연구들이 선행되어 왔다. 본 연구에서는 웹사이트 Kaggle에서 제공한 48*48 8-bit grayscale 이미지 데이터셋을 사용하여 얼굴인식과 표정분류로 구분된 두 단계를 거치는 얼굴표정 자동 인식 시스템을 구축하였고, 이를 기존의 연구와 비교하여 자료 및 방법론의 특징을 고찰하였다. 분석 결과, Face landmark 정보에 주성분분석을 적용하여 단 30개의 주성분만으로도 빠르고 효율적인 예측모형을 얻을 수 있음이 밝혀졌다. LDA, Random forest, SVM, Bagging 중 SVM방법을 적용했을 때 가장 높은 정확도를 보이며, LDA방법을 적용하는 경우는 SVM 다음으로 높은 정확도를 보이며, 매우 빠르게 적합하고 예측하는 것이 가능하다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제8권9호
• /
• pp.379-390
• /
• 2019

본 논문에서는 시각-언어 이동 문제를 위한 새로운 심층 신경망 모델인 LVLN을 제안한다. LVLN 모델에서는 자연어 지시의 언어적 특징과 입력 영상 전체의 시각적 특징들 외에, 자연어 지시에서 언급하는 주요 장소와 랜드마크 물체들을 입력 영상에서 탐지해내고 이 정보들을 추가적으로 이용한다. 또한 이 모델은 자연어 지시 내 각 개체와 영상 내 각 관심 영역, 그리고 영상에서 탐지된 개별 물체 및 장소 간의 서로 연관성을 높일 수 있도록 맥락 정보 기반의 주의 집중 메커니즘을 이용한다. 그뿐만 아니라, LVLN 모델은 에이전트의 목표 도달 성공율을 향상시키기 위해, 목표를 향한 실질적인 접근을 점검할 수 있는 진척 점검기 모듈도 포함하고 있다. Matterport3D 시뮬레이터와 Room-to-Room (R2R) 벤치마크 데이터 집합을 이용한 다양한 실험들을 통해, 본 논문에서 제안하는 LVLN 모델의 높은 성능을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• Journal of International Society for Simulation Surgery
• /
• 제2권1호
• /
• pp.26-32
• /
• 2015

Purpose When a surgeon examines the morphology of skull of patient, locations of craniometric landmarks of 3D computed tomography(CT) volume are one of the most important information for surgical purpose. The locations of craniometric landmarks can be found manually by surgeon from the 3D rendered volume or 2D sagittal, axial, and coronal slices which are taken by CT. Since there are many landmarks on the skull, finding these manually is time-consuming, exhaustive, and occasionally inexact. These inefficiencies raise a demand for a automatic localization technique for craniometric landmark points. So in this paper, we propose a novel method through which we can automatically find these landmark points, which are useful for surgical purpose. Materials and Methods At first, we align the experimental data (CT volumes) using Frankfurt Horizontal Plane (FHP) and Mid Sagittal Plane(MSP) which are defined by 3 and 2 cranial landmark points each. The target landmark of our experiment is the anterior nasal spine. Prior to constructing a statistical cubic model which would be used for detecting the location of the landmark from a given CT volume, reference points for the anterior nasal spine were manually chosen by a surgeon from several CT volume sets. The statistical cubic model is constructed by calculating weighted intensity means of these CT sets around the reference points. By finding the location where similarity function (squared difference function) has the minimal value with this model, the location of the landmark can be found from any given CT volume. Results In this paper, we used 5 CT volumes to construct the statistical cubic model. The 20 CT volumes including the volumes, which were used to construct the model, were used for testing. The range of age of subjects is up to 2 years (24 months) old. The found points of each data are almost close to the reference point which were manually chosen by surgeon. Also it has been seen that the similarity function always has the global minimum at the detection point. Conclusion Through the experiment, we have seen the proposed method shows the outstanding performance in searching the landmark point. This algorithm would make surgeons efficiently work with morphological informations of skull. We also expect the potential of our algorithm for searching the anatomic landmarks not only cranial landmarks.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2003년도 추계종합학술대회 논문집
• /
• pp.423-430
• /
• 2003

Generally, elements of DRM(Digital Road Map) consist of road, background and landmark data. The landmark, expressed as text and symbol, on map and additional search data are processed by GISSD(Geo-spatial Information System Service Data). This paper aims to develop the DBMS(Database Management System) for operating landmark and search data, and to discuss the characteristics and application of the DBMS. To accomplish the two objectives, the following four tasks were performed in this study. First, the working scopes of field survey and specification to construct the GISSD were defined. Second, the suggested process of manufacture and design of database were described. Third, the software for required construction and management of the system were developed. Lastly, the properties of developed system and data were analyzed. Especially, the efforts for the GISSD in this study are expected to provide a direct use and practical application to the creation of landmark in DRM and search data.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.178-180
• /
• 2022

본 논문은 Java 기반의 Firebase용 ML Kit로부터 구한 요가 강사의 요가 자세의 랜드마크를 기반으로 사용자가 요가 자세 훈련할 수 있는 애플리케이션 구현을 소개한다. ML Kit를 이용해 사용자의 자세를 분류하고 각 관절에 해당하는 랜드마크를 구한다. 구해진 랜드마크의 관절들이 이루는 각도를 통해 요가 자세에 대한 정확도 측정 기준값을 설정한다. 전문 요가 강사의 요가 자세에 대한 기준 랜드마크와 ML Kit를 통한 사용자의 포즈에 대한 랜드마크 사이의 정확도를 비교한다. 정확도 기준값에 따라서 오동작과 정동작 정보를 TTS(Text-to-Speech)를 통해 사용자에게 제공해준다. Firebase로 효과적인 사용자관리를 하고, 사용자 요가 자세가 정확도 기준값에 부합하는 운동을 했을 경우에 카운터와 타이머를 통해 운동량을 디스플레이하는 시스템을 설명한다.

• 대한인간공학회지
• /
• 제23권1호
• /
• pp.23-38
• /
• 2004

This study examines different types of navigation aids when a navigator performs target search tasks in Virtual Environments. The factors manipulated in this study include target information (None/Landmark). navigational difficulty (Easy/Difficult). and map types (None/2D Map/3D Map). Navigation performance was measured by using task completion time and the number of target locations that was remembered by the navigator. In addition, user satisfaction on the navigation aids was also measured by using a 7-point Likert's scale. The results showed that the user satisfaction on the landmark was high when the 3D Map was provided. The task completion time shortened when navigational difficulty was set at "easy." The number of remembered target locations was large when there was no landmark. It was also large with an easy navigation task. or a map (20 or 3D) provided. Guidelines for selecting navigation aids were proposed based on the results.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1990년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 26-27 Oct. 1990
• /
• pp.716-721
• /
• 1990

It is one of the essential task to determine the absolute location of mobile robot during its navigation. In this paper we propose an algorithm to calculate the distance and orientation of camera from landmark through the visual image of stripe typed landmark. Exact closed form solution of camera location is obtained with the correspondences from vertical line on mark plane to the intersection point of projected line with horizontal axis of image plane. It needs only one line image information, so that location determination can be processed in real time.