• International Journal of Aeronautical and Space Sciences
• /
• 제13권1호
• /
• pp.99-105
• /
• 2012

Track-before-detect techniques based on dynamic programming have provided solutions for detecting targets from a sequence of images. In its application to airborne threat detection, dynamic programming solutions should take into account the distinguishable properties of objects in a collision course. This paper describes the development of a new track scoring function that accumulates scores for airborne targets in Bayesian framework. Numerical results show that the proposed scoring function has slightly better detection capabilities.

• 대한임베디드공학회논문지
• /
• 제6권5호
• /
• pp.319-325
• /
• 2011

Recently, collision avoidance systems are under development to reduce the traffic accidents and driver comfort for automotive radar. Pulse radar can detect their range and velocities of moving vehicles using range gate and FFT(Fast Fourier Transform) of the doppler frequency. We designed the real time DSP(Digital Signal Processing) based automotive UWB(Ultra Wideband) radar, and implemented DSP to detect the range and velocity within 100ms for real time system of the automotive UWB radar. We also measured the range and velocity of a moving vehicle using designed automotive UWB radar in a real road environment.

• 한국정보전자통신기술학회논문지
• /
• 제10권4호
• /
• pp.318-324
• /
• 2017

본 항공기에 스테레오 카메라를 장착하여 영상 기반의 비행 객체 탐지 및 탐지된 객체의 3차원 위치를 추정하는 방법을 제시하였다. 구름 사이에 존재할 수 있는 원거리의 작은 객체를 탐지하기 위한 방법으로 PCT 기반의 Saliency Map을 생성하여 이용하였으며, 이렇게 탐지된 객체는 좌우 스테레오 영상에서 매칭을 수행하여 스테레오 시차(Disparity)를 추출하였다. 정확한 Disparity를 추출하기 위하여 비용집적(Cost Aggregation) 영역을 탐지 객체에 맞추어 가변되도록 가변 영역으로 사용하였으며, 본 논문에서는 Saliency Map에서 객체의 존재 영역으로 검출된 결과를 사용하였다. 좀 더 정밀한 Disparity를 추출하기 위하여 Sub-pixel interpolation 기법을 사용하여 Sub-pixel 레벨의 실수형 Disparity를 추출하였다. 또한 이에 카메라 파라미터를 적용하여 실제 탐지된 비행 객체의 3차원 공간 좌표를 생성하여 객체의 공간위치를 추정하는 방법을 제시하였다. 이는 향후 자율비행체의 영상기반 객체 탐지 및 충돌방지 시스템에 활용될 수 있을 것으로 기대된다.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.1243-1244
• /
• 2022

In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

• 한국항공우주학회지
• /
• 제39권9호
• /
• pp.848-857
• /
• 2011

영상센서 기반의 충돌회피 시스템을 구성하기 위해서는 수 픽셀 이내의 낮은 신호대잡음비 환경에서 다수의 표적을 탐지할 수 있는 알고리듬이 필요하다. 이처럼 영상 내에서 희미하게 나타나는 잠재적인 표적과 잡음을 구분하기 위한 방법으로서 연속적인 영상 정보를 효율적으로 처리하는 Track-Before-Detect (TBD) 알고리듬이 연구되고 있다. 본 논문에서는 기존의 TBD 알고리듬을 확장하여 다수 표적 탐지 요구조건을 만족시키기 위한 두 가지 방식의 기법을 제시하였다. 첫 번째 방식은 동적 계획법과 K-평균 클러스터링 기법에 기반을 두고 있으며 두 번째 방식은 은닉 마르코프 모델에 Sub-Window 기법을 적용하였다. 제안한 방식의 성능 및 차이점은 수치해석 결과를 통해 분석하였다.

• 한국인터넷방송통신학회논문지
• /
• 제18권3호
• /
• pp.35-41
• /
• 2018

본 논문에서는 햅틱 렌더링을 위한 동영상의 장면을 정의하고 그 중 정지 상태의 객체와 단일 객체가 충돌하는 상황에 대한 햅틱 렌더링 알고리즘을 제안한다. 객체의 움직임을 크게 네 가지 상황으로 분류하였으며 동영상에서 쉽게 볼 수 있는 장면들을 활용하였다. 제안하는 알고리즘은 움직임 예측과 객체 추적을 활용하여 객체의 방향 변화와 모양 변화를 검출함으로써 햅틱 렌더링을 하기에 적합한 프레임인지 판별할 수 있도록 하였다. 알고리즘 실행을 통해 샘플 동영상에서 총 13개의 프레임을 추출하였으며 해당 프레임이 재생되는 시간을 계산하였다. 햅틱 편집 프로그램을 통해 상황에 알맞은 햅틱 생성 파형을 추가함으로써 계산된 재생 시간에 햅틱 효과가 나타나는 것을 확인하였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권4호
• /
• pp.316-321
• /
• 2005

We designed and implemented fish robots for various purposes such as autonomous navigation, maneuverability control, posture balancing and improvement of quick turns in a tank of 120 X 120 X 180cm size. Typically, fish robots have 30-50 X 15-25 X 10-20cm dimensions; length, width and height, respectively. It is essential to have the ability of quick and smooth turning to avoid collision with obstacles or walls of the water pool at a close distance. Infrared distance sensors are used to detect obstacles, magneto-resistive sensors are used to read direction information, and a two-axis accelerometer is mounted to compensate output of direction sensors. Because of the swing action of its head due to the tail fin movement, the outputs of an infrared distance sensor contain a huge amount of noise around true distances. With the information from accelerometers and e-compass, much improved distance data can be obtained by fuzzy logic based estimation. Successful swimming and smooth turns without collision demonstrated the effectiveness of the distance estimation.

• International Journal of Computer Science & Network Security
• /
• 제23권9호
• /
• pp.141-149
• /
• 2023

Repacked mobile apps constitute about 78% of all malware of Android, and it greatly affects the technical ecosystem of Android. Although many methods exist for repacked app detection, most of them suffer from performance issues. In this manuscript, a novel method using the Constant Key Point Selection and Limited Binary Pattern (CKPS: LBP) Feature extraction-based Hashing is proposed for the identification of repacked android applications through the visual similarity, which is a notable feature of repacked applications. The results from the experiment prove that the proposed method can effectively detect the apps that are similar visually even that are even under the double fold content manipulations. From the experimental analysis, it proved that the proposed CKPS: LBP method has a better efficiency of detecting 1354 similar applications from a repository of 95124 applications and also the computational time was 0.91 seconds within which a user could get the decision of whether the app repacked. The overall efficiency of the proposed algorithm is 41% greater than the average of other methods, and the time complexity is found to have been reduced by 31%. The collision probability of the Hashes was 41% better than the average value of the other state of the art methods.

• 한국콘텐츠학회논문지
• /
• 제6권1호
• /
• pp.72-79
• /
• 2006

최근의 MMORPG 게임은 심리스 월드로 분할하는 분산 서버를 구축하고 있다. 본 연구는 이러한 분산서버 간의 공유영역에 해당되는 지역에 대한 충돌처리를 다룬다. 분산 서버간의 공유 영역에 대한 경계영역을 동적으로 조정하기 위해 DLS을 사용하고, 광선과 단말 노드 간의 충돌 위치 관계를 통하여 이웃 노드를 빠르게 탐색한다. 이렇게 구해진 노드의 값을 통하여 객체 간의 충돌처리를 판별한다. 이것은 각 서버가 공유 영역에 대한 정보를 계속 보유할 필요가 없고, 서버간의 경계 영역을 포인터를 이용하여 빠르게 탐색할 수 있게 한다. 충돌은 계층적 경계상자를 이용하여 인접한 개체의 값들을 그룹으로 이진트리로 구축한다. 이러한 처리 방법은 처리량을 이분화 시켜 효과적으로 처리량을 줄일 수 있다. 또한 실시간적으로 발생되는 개체 정보의 변경 시 공유영역에 대한 복사가 필요하지 않으므로 네트워크 트래픽에 대한 처리량도 효과적으로 줄일 수 있다.

• 한국정밀공학회지
• /
• 제17권7호
• /
• pp.113-123
• /
• 2000

In the navigation for a mobile robot, the collision avoidance with unexpected obstacles is essential for the safe navigation and it is independent of the technique used to control the mobile robot. This paper presents a new collision avoidance algorithm using neural network for the safe navigation of the autonomous mobile robot equipped with CAN and ultrasonic sensors. A tracked wheeled mobile robot has a stability and an efficiency to move on a rough ground. And its mechanism is simple. However it has difficulties to recognize its surroundings. Because the shape of the tracked wheeled mobile robot is a square type, sensor modules are generally located on the each plane surface of 4 sides only. In this paper, the algorithm using neural network is proposed in order to avoid unexpected obstacles. The important character of the proposed algorithm is to be able to detect the distance and the angle of inclination of obstacles. Only using datum of the distance and the angle, informations about the location and shape of obstacles are obtained, and then the driving direction is changed. Consequently, this algorithm is capable of real time processing and available for a mobile robot which has few sensor modules or the limited sensing range such as a tracked wheeled mobile robot. Effectiveness of the proposed algorithm is illustrated through a computer simulation and an experiment using a real robot.