Earticle

There have been continuous efforts on automating welding processes. This automation process could he said to fall into two categories, weld seam tracking and weld quality evaluation. Recently, the attempts to achieve these two functions simultaneously arc on the increase. For the study presented in this paper, a vision sensor is made, a vision system is constructed and using this, the 3 dimensional geometry of the head is measured on-line. For the application as in welding, which is the characteristic of nonlinear process, a fuzzy controller is designed. And with this, an adaptive control system is proposed which acquires the bead height and the coordinates of the point on the bead along the horizontal fillet joint, performs seam tracking with those data, and also at the same time, controls the bead geometry to a uniform shape. A communication system, which enables the communication with the industrial robot, is designed to control the bead geometry and to track the weld seam. Experiments are made with varied offset angles from the pre-taught weld path, and they showed the adaptive system works favorable results.

This paper presents a method of measuring the 6 DOF　motion of a micro object using images taken of interference fringes projected onto the object. Information from the fringe patterns allows for extracting the 6 DOF motion of the object in one image, allowing for real time measurement of the object's pose. This measurement technique is applied to a visual servo control scheme where the object's 6 DOF　motion is controled. Experimental results of the developed system are presented.

We propose a novel programmable miniature vision module based on a custom designed analog VLSI (aVLSI) chip. The vision module consists of the optical flow vision sensor embedded with commercial off-the-shelves digital hardware; in our case is the Intel XScale PXA270 processor enforced with a programmable gate array device. The aVLSI sensor provides gray-scale imager data as well as smooth optical flow estimates, thus each pixel gives a triplet of information that can be continuously read out as three independent images. The particular computational architecture of the custom designed sensor, which is fully parallel and also analog, allows for efficient real-time estimations of the smooth optical flow. The Intel XScale PXA270 controls the sensor read-out and furthermore allows, together with the programmable gate array, for additional higher level processing of the intensity image and optical flow data. It also provides the necessary standard interface such that the module can be easily programmed and integrated into different vision systems, or even form a complete stand-alone vision system itself. The low power consumption, small size and flexible interface of the proposed vision module suggests that it could be particularly well suited as a vision system in an autonomous robotics platform and especially well suited for educational projects in the robotic sciences.

최근 극단적인 강우와 홍수의 발생 빈도가 증가함에 따라, 수자원 시설은 보다 강하고 빈번한 환경 스트레스에 노출되고 있다. 이처럼 반복적인 환경 스트레스는 구조물의 균열 형성과 확산을 가속화하며, 내구성 저하와 안전성 문제로 이어진다. 이에 본 연구에서는 건물 균열 이미지 데이터를 기반으로, 다단계 인공지능 영상 인식 가상센서를 적용한 수자원 시설 균열 자동 감지 기법을 제안하고자 한다. 첫째, YOLO11 기반 벽체 영역분할을 통해 관심 영역을 설정하고, 둘째, 동일 계열의 균열 탐지 모델로 균열 후보를 검출한다. 셋째, ResNet18 기반 이진 분류로 실제 균열 여부를 판별하며. 마지막으로 YOLO11 기반 세분화(Segmentation)모델을 적용하여 균열의 형상을 정밀하게 인식한다. 본 연구의 결과는 계층적 인공지능 분석 체계를 통해 수자원 시설 구조물의 균열을 효율적으로 감지하고 손상 위험을 조기에 파악할 수 있는 과학적 근거를 마련하는 데 기여할 것으로 기대된다.

With the increasing frequency of extreme rainfall and flooding events, water infrastructure has been increasingly exposed to stronger and more frequent environments stresses. Such repetitive external loads accelerate the formation and propagation of cracks in structures, leading to reduced durability and potential safety issues. In this study, a multi-stage artificial intelligence vision-based virtual sensor approach is proposed to automatically identify cracks in water infrastructure using building crack image data. First, the region of interest (ROI) is defined through wall segmentation based on YOLO11, followed by crack candidate detection using the same YOLO11 framework. Next, a ResNet18-based binary classification model determines the actual presence of cracks, and finally, a YOLOv11-based segmentation model is applied to precisely identify the crack geometry. The Proposed multi-stage AI analysis framework enables efficient crack identify in water infrastructure and provides a scientific foundation for the early identification of potential structural damage.

본 논문은 레이더 센서, 비전 센서를 활용한 다중 센서 융합 기반 움직임 검지에 관한 연구를 다룬다. 레이더 센서는 다량의 물체를 검지함에 있어 센서 자체의 움직임이 발생할 경우 주변건물이나 주변 가로수와 같은 사물 혹은 물체를 차량으로 오인하는 경우가 생긴다. 비전 센서의 경우 저렴하고 가장 많이 쓰는 형태이지만 빛, 흔들림, 날씨, 조도 등 외부환경에 취약하다는 문제점이 있다. 각 센서 간의 문제점을 보완하고자 센서 융합을 통한 움직임검지를 제안하게 되었고 실험환경 내에서 매우 우수한 검지율을 보이게 되었다 센서 간 융합에서 좌표 통일문제와 실시간 전송문제 등을 해결하였으며, 각 센서 간 필터링을 통한 비가공데이터(raw data)의 신뢰성을 높였다. 특히 영상에서는 가우시안 혼합모델(GMM, Gaussian Mixture Model)을 사용하여 레이더 센서의 단점을 극복하고자 했다.

This Paper is for A study of sensor fusion using Radar sensor and Vision sensor in moving object detection. Radar sensor has some problems to detect object. When the sensor moves by wind or that kind of thing, it can happen to detect wrong object like building or tress. And vision sensor is very useful for all area. And it is also used so much. but there are some weakness that is influenced easily by the light of the area, shaking of the sensor device, and weather and so on. So in this paper I want to suggest to fuse these sensor to detect object. Each sensor can fill the other’s weakness, so this kind of sensor fusion makes object detection much powerful.

본 논문은 위성통신 기반의 위성 영상감시 센서 네트워크 적용을 위한 스마트 비젼 센서에 대해 기술한다. 스마트 비젼센서 단말은 현장에서 산불, 연기, 침입자 움직임 등의 이벤트를 자동감지하면서 높은 성능 신뢰도, 견고한 하드웨어 내구성, 용이한 유지보수, 끊김없는 통신유지 기능들이 요구된다. 이러한 요구사항들을 만족시키기 위하여 스마트 비젼 센서가 내장된 초소형 위성통신 단말 을 제안하며 위성 송수신 기능과 더불어 고 신뢰도의 임베디드 영상분석 및 영상압축 기능을 처리한다. 제안하는 비젼 센서 알고리 즘의 컴퓨터 시뮬레이션과 비젼 센서 시제품 시험을 통하여 영상감시 성능을 검증하였으며 실용성을 확인하였다.

In this paper, satellite communication based video surveillance system that consisted of ultra-small aperture terminals with small-size smart vision sensor is proposed. The events such as forest fire, smoke, intruder movement are detected automatically in field and false alarms are minimized by using intelligent and high-reliable video analysis algorithms. The smart vision sensor is necessary to achieve high-confidence, high hardware endurance, seamless communication and easy maintenance requirements. To satisfy these requirements, real-time digital signal processor, camera module and satellite transceiver are integrated as a smart vision sensor-based ultra-small aperture terminal. Also, high-performance video analysis and image coding algorithms are embedded. The video analysis functions and performances were verified and confirmed practicality through computer simulation and vision sensor prototype test.

최근 기후변화, 교통혼잡, 범죄예방 등 여러 사회 문제 해결을 위한 드론의 필요성이 증가함에 따라 엔터테인먼트, 에너지, 건설, 교통, 농업 등 다양한 산업과 여러 사회 서비스로 그 필요성이 확대되고 있다. 또한, 4차 산업혁명에 맞춰 인공지능과의 융·복합을 통한 드론의 자율 착륙 연구는 주요한 과제이기도 하다. 본 연구는 이러한 사회적 흐름 에 따라 소형 드론을 활용하여 인공지능 기술을 통한 드론의 활용성을 확대하고 GPS 수신이 안 되는 영역에서 딥러 닝 객체 탐지 모델을 활용한 자율 착륙을 연구하였다. GPS 수신기를 통해 받는 위치 정보 대신 드론에 장착된 카메 라를 통해 전달받는 영상에서 착륙할 지점을 인식하고 카메라를 통해 받는 영상 정보만 이용하여 목표지점에 하강하 는 방식으로 자율 착륙을 유도하였다. 딥러닝 중 경량화 모델의 대표적인 SSD-MobileNet v2와 YOLOv4-tiny 모 델을 사용하였고, 효율적으로 착륙 지점을 인식하기 위해 최적화 과정을 진행해 실시간 자율 착륙이 가능하게 했다. 실내에서의 비행시험을 통해 GPS 음영지역에서 사람의 조종이 없이 드론을 안전하게 착륙을 시킬 수 있음을 확인 하였다.

Recently, as the need for drones to solve various social problems such as climate change, traffic congestion, and crime prevention increases, the need for drones is expanding to various industries and social services such as entertainment, energy, construction, transportation, and agriculture. In addition, research on autonomous landing of drones through convergence and integration with artificial intelligence in line with the 4th industrial revolution is also a major task. In accordance with this social trend, this study expanded the use of drones through artificial intelligence technology by using small drones and studied autonomous landing using deep learning object detection models in areas where GPS receivers are not available. In areas where GPS signal reception is difficult, autonomous landing is induced by recognizing the landing point from the image transmitted through the camera mounted on the drone instead of the location information received through the GPS receiver and descending to the target point using only the image information. Using SSD-MobileNet v2 and YOLOv4-tiny models, which are representative lightweight models during deep learning an optimization process was performed to detect an effective landing point, enabling real-time autonomous landing. Through this study, it is expected that the dependence on the GPS receiver and human control for drone landing can be reduced.

본 연구에서는 개별 차량의 차두 시간(time headway) 정보를 활용하여 고속도로 환경에서의 단일 링크에 대한 교통류 모형(flow-density diagram)을 추정하는 방법에 대해 탐구한다. 차두 시간 기반 교통류 모형(empirical flow-density diagram) 연구를 위해 차량용 비전 센서가 탑재된 실험 차량에서 9개월동안 수집된 데이터의 전처리 및 GIS 기반 맵 매칭을 수행한다. 기존의 교통류 모델식을 활용한 차두 시간 기반 교통류 모형(empirical flow-density diagram)의 검증을 위해, 차량 검지기 기반의 VDS(Vehicle Detection System) 데이터(loop detection traffic data) 기반 교통류 모형과 결과 비교 및 분석을 수행한다. 차두 시간 기반 교통류 모형의 추정 오차 원인 을 분석하기 위해 각 교통류 모형의 차두 시간 및 차두 거리의 확률분포와 단위시간 교통량과 차량 밀도의 표준편차를 활용하였다. 분석 결과 링크 내 제한된 샘플 차량 대수 및 수집 데이 터에 대한 주행환경 편향성이 추정 오차의 주된 요인이며. 이에 따른 추정 오차 개선을 위한 방법에 대해 제안한다.

This study explored an approach to estimate a flow-density diagram(FD) on a link in highway traffic environment by utilizing probe vehicles’ time headway records. To study empirical flow-density diagram(EFD), the probe vehicles with vision sensors were recruited for collecting driving records for nine months and the vision sensor data pre-processing and GIS-based map matching were implemented. Then, we examined the new EFDs to evaluate validity with reference diagrams which is derived from loop detection traffic data. The probability distributions of time headway and distance headway as well as standard deviation of flow and density were utilized in examination. As a result, it turned out that the main factors for estimation errors are the limited number of probe vehicles and bias of flow status. We finally suggest a method to improve the accuracy of EFD model.

Measurement of surface roughness is an important metrology procedure widely used in the industry during the final phases of the manufacturing process for quality control. This paper presents a measurement system that uses an optical microscope equipped with a vision sensor to measure the surface roughness of bearings. Shape from Focus / Defocus method is applied to measure the 3D surface coordinates of the object surface from which roughness calculations can be undertaken.

기존의 첨단 운전자 지원 시스템 (Advanced Driver Assistance System, ADAS)들은 전방 위험 탐지와 같은 한정적 상황에서의 사고 예방에 집중하고 있어 다양한 사고 시나리오가 존재하는 교차로에 적용하기에는 한계를 가지고 있다. 또한 기존 연구는 주로 사고 요인 분석에 집중하고 있어 첨단 운전자 지원 시스템의 사고 예방 성능에 관한 연구는 미비한 편이다. 이에 본 연구에서 는 비전 및 레이더 센서 기반 첨단 운전자 지원 시스템의 다양한 교차로 사고 예방에 대한 성능을 평가하고 대책을 마련하고자 한다. 이를 위하여 미국의 Second Strategic Highway Research Program(SHRP2)의 naturalistic driving study(NDS)에서 수집된 사고/준사고 상황의 거리 측정 데이 터를 기반으로 16개의 교차로 사고 시나리오를 도출하였고, 총 363건의 차량과 차량 간 사고를 분석하였다. 분석 결과 16개의 사고 시나리오 중 0.7의 사고 예방율을 기준으로 카메라 기반 시스템은 5개, 레이더 기반 시스템은 4개의 사고 시나리오에서 사고를 예방할 수 있었다.

The current collision warning and avoidance system(CWAS) is one of the representative Advanced Driver Assistance Systems (ADAS) that significantly contributes to improve the safety performance of a vehicle and mitigate the severity of an accident. However, current CWAS mainly have focused on preventing a forward collision in an uninterrupted flow, and the prevention performance near intersections and other various types of accident scenarios are not extensively studied. In this paper, the safety performance of Vision-Sensor (VS) and Radar-Sensor(RS)–based collision warning systems are evaluated near an intersection area with the data from Naturalistic Driving Study(NDS) of Second Strategic Highway Research Program(SHRP2). Based on the VS and RS data, we newly derived sixteen vehicle-to-vehicle accident scenarios near an intersection. Then, we evaluated the detection performance of VS and RS within the derived scenarios. The results showed that VS and RS can prevent an accident in limited situations due to their restrained field-of-view. With an accident prevention rate of 0.7, VS and RS can prevent an accident in five and four scenarios, respectively. For an efficient accident prevention, a different system that can detect vehicles’movement with longer range than VS and RS is required as well as an algorithm that can predict the future movement of other vehicles. In order to further improve the safety performance of CWAS near intersection areas, a communication-based collision warning system such as integration algorithm of data from infrastructure and in-vehicle sensor shall be developed.

In this paper, we study on aerial objects detection and position estimation algorithm for the safety of UAV that flight in BVLOS. We use the vision sensor and LiDAR to detect objects. We use YOLOv2 architecture based on CNN to detect objects on a 2D image. Additionally we use a clustering method to detect objects on point cloud data acquired from LiDAR. When a single sensor used, detection rate can be degraded in a specific situation depending on the characteristics of sensor. If the result of the detection algorithm using a single sensor is absent or false, we need to complement the detection accuracy. In order to complement the accuracy of detection algorithm based on a single sensor, we use the Kalman filter. And we fused the results of a single sensor to improve detection accuracy. We estimate the 3D position of the object using the pixel position of the object and distance measured to LiDAR. We verified the performance of proposed fusion algorithm by performing the simulation using the Gazebo simulator.

The role of the library is as a public institution that provides academic information to a variety of people, including students, the general public, and researchers. These days, as the importance of lifelong education is emphasized, libraries are evolving beyond simply storing and lending materials to complex cultural spaces that share knowledge and information through various educational programs and cultural events. One of the problems library user’s faces is locating books to borrow. This problem occurs because of errors in the location of borrowed books due to delays in updating library databases related to borrowed books, incorrect labeling, and books temporarily located in different locations. The biggest problem is that it takes a long time for users to search for the books they want to borrow. In this paper, we propose a system that visually displays the location of books in real time using an AI vision sensor and LED. The AI vision sensor-based book location guidance system generates a QR code containing the call number of the borrowed book. When the AI vision sensor recognizes this QR code, the exact location of the book is visually displayed through LED to guide users to find it easily. We believe that the AI vision sensor-based book location guidance system dramatically improves book search and management efficiency, and this technology is expected to have great potential for use not only in libraries and bookstores but also in a variety of other fields.