• 방송공학회논문지
• /
• 제28권1호
• /
• pp.124-131
• /
• 2023

원거리에서 특정 영역의 물리적 특성 또는 상황에 대한 정보를 얻기 위해 원격 탐사 영상에 객체 검출 기법이 연구되고 있다. 이때 저해상도인 원격 영상은 정보의 손실로 인해 객체 검출의 정확도가 떨어지는 문제가 발생한다. 본 논문에서는 이러한 문제점을 해결하기 위해 초고해상도 기법과 객체 검출 방법을 하나의 네트워크로 구성하여 원격 영상에서 객체 검출의 성능을 높이는 방법을 제안한다. 제안한 방법은 심층 잔차 밀집 기반의 네트워크를 구성하여 저해상도 영상에서 객체의 특징을 복원하고자 하였다. 추가적으로 이를 객체 검출 단계인 YOLOv5와 하나의 네트워크로 구성함으로써 객체 검출의 성능을 향상시키고자 하였다. 제안한 방법은 저해상도 영상을 위해 VEDAI 데이터를 이용하였으며 차량 검출에서 VISIBLE 기준으로 mAP@0.5에 대해 81.38%까지 향상됨을 확인하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.487-490
• /
• 2003

In this research, we will devise an obstacle avoidance algorithm for a previously unmanned vehicle. Whole systems consist mainly of the vehicle system and the control system. The two systems are separated; this system can communicate with the vehicle system and the control system through wireless RF (Radio Frequency) modules. These modules use wireless communication. And the vehicle system is operated on PIC Micro Controller. Obstacle avoidance method for unmanned vehicle is based on the Virtual Force Field (VFF) method. An obstacle exerts repulsive forces and the lane center point applies an attractive force to the unmanned vehicle. A resultant force vector, comprising of the sum of a target directed attractive force and repulsive forces from an obstacle, is calculated for a given unmanned vehicle position. With resultant force acting on the unmanned vehicle, the vehicle's new driving direction is calculated, the vehicle makes steering adjustments, and this algorithm is repeated.

• International Journal of Automotive Technology
• /
• 제7권1호
• /
• pp.83-90
• /
• 2006

The purpose of this paper is to describe the design and implementation of an unmanned ground vehicle, called the TailGator at CIMAR (Center for Intelligent Machines and Robotics) of the University of Florida. The TailGator is a gas powered, four-wheeled vehicle that was designed for the AUVSI Intelligent Ground Vehicle Competition and has been tested in the contest for 2 years. The vehicle control model and design of the sensory systems are described. The competition is comprised of two events called the Autonomous Challenge and the Navigation Challenge: For the autonomous challenge, line following, obstacle avoidance, and detection are required. Line following is accomplished with a camera system. Obstacle avoidance and detection are accomplished with a laser scanner. For the navigation challenge, waypoint following and obstacle detection are required. The waypoint navigation is implemented with a global positioning system. The TailGator has provided an educational test bed for not only the contest requirements but also other studies in developing artificial intelligence algorithms such as adaptive control, creative control, automatic calibration, and internet-base control. The significance of this effort is in helping engineering and technology students understand the transition from theory to practice.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권10호
• /
• pp.955-961
• /
• 2010

본 논문에서는 HOG 특정벡터와 영상분할을 이용한 부스팅 분류기반의 자동차영역 검출 알고리즘의 연구에 대해서 기술한다. 입력된 영상으로부터 차량을 검출하기위해 먼저 분할 후 합병(split-merge) 방법을 적용하여 영상을 분할한다. 그리고 가장 큰 두 영역을 검색 영역에서 제외하여 처리 속도를 향상 시킨다. 각 영역에 대해 HOG(histogram of oriented gradient) 특정을 추출한다. 분류기는 두 개의 모집단을 분류하는데 많이 사용되고 있는 AdaBoost 방법을 사용한다. 제안방법의 성능 평가를 위해 537개의 영상을 사용하여 분류기를 학습하였으며, 또한 학습에 사용하지 않은 비학습영상 500개를 사용하여 인식률을 구하였다. 실험결과 비학습영상에 대해 98.34%의 인식률을 얻었다. 결론적으로 제안된 방법이 지능형 자동차 제어 시스템에서 차량의 위치를 찾는 방법으로 활용될 수 있다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.159-162
• /
• 2000

Selecting locally optimum thresholds, based on optimizing a criterion composed of the area variation rate and the compactness of the segmented shape, is presented. The method is shown to have the shape-resolving property in the subtraction image, so that overlapped objects may be resolved into bright and dark evidences characterizing each object. As an application a vehicle detection algorithm robust to the operating conditions could be realized by applying simple merging rules to the geometrically correlated bright and dark evidences obtained by this local thresholding.

• 제어로봇시스템학회논문지
• /
• 제17권8호
• /
• pp.799-806
• /
• 2011

This paper proposes an algorithm to detect the stop line and crosswalk on the road surface using edge information and blob coloring. The detection has been considered as an important area of autonomous vehicle technologies. The proposed algorithm is composed of three phases: 1) hypothesis generation of stop lines, 2) hypothesis generation of crosswalks, and 3) hypothesis verification of stop lines. The last two phases are not performed if the first phase does not provide a hypothesis of a stop line. The last one is carried out by the combination of both hypotheses of stop lines and crosswalks, and determines the stop lines among stop line hypotheses. The proposed algorithm is proven to be effective through experiments with various images captured on the roads.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권7호
• /
• pp.465-472
• /
• 2004

Analog CPPN-based optimal road boundary detection algorithm for autonomous vehicle is proposed. The CPPN is a massively connected analog parallel array processor. In the paper, the dynamic programming which is an efficient algorithm to find the optimal path is implemented with the CPPN algorithm. If the image of road-boundary information is utilized as an inter-cell distance, and goals and start lines are positioned at the top and the bottom of the image, respectively, the optimal path finding algorithm can be exploited for optimal road boundary detection. By virtue of the parallel and analog processing of the CPPN and the optimal solution of the dynamic programming, the proposed road boundary detection algorithm is expected to have very high speed and robust processing if it is implemented into circuits. The proposed road boundary algorithm is described and simulation results are reported.

• Journal of electromagnetic engineering and science
• /
• 제19권1호
• /
• pp.13-19
• /
• 2019

This paper proposes a systematic method for estimating detection performances of a frequency-modulated continuous wave radar using electromagnetic simulations. The proposed systematic method includes a radar system simulator that can obtain range-Doppler images using the electromagnetic (EM) simulations in conjunction with a test setup employed for performance evaluation of multiple targets at different velocities in a traffic environment. This method is then applied for optimizing the half-power beamwidths of the antenna array using an evaluation metric defined to improve the detection strengths for the multiple targets. The optimized antenna has vertical and horizontal half-power beam widths of $10^{\circ}$ and $60^{\circ}$, respectively. The results confirm that that the proposed systematic method is suitable to improve the radar detection performance with the enhanced radar-Doppler images.

• 제어로봇시스템학회논문지
• /
• 제21권8호
• /
• pp.766-772
• /
• 2015

This paper proposes an object detection method based on motion estimation using background subtraction in the fisheye images obtained through omni-directional camera mounted on the vehicle. Recently, most of the vehicles installed with rear camera as a standard option, as well as various camera systems for safety. However, differently from the conventional object detection using the image obtained from the camera, the embedded system installed in the vehicle is difficult to apply a complicated algorithm because of its inherent low processing performance. In general, the embedded system needs system-dependent algorithm because it has lower processing performance than the computer. In this paper, the location of object is estimated from the information of object's motion obtained by applying a background subtraction method which compares the previous frames with the current ones. The real-time detection performance of the proposed method for object detection is verified experimentally on embedded board by comparing the proposed algorithm with the object detection based on LKOF (Lucas-Kanade optical flow).

• 제어로봇시스템학회논문지
• /
• 제15권2호
• /
• pp.155-162
• /
• 2009

In walking guide robot, a guide vehicle detects an obstacle distribution in the walking space using range sensors, and generates a 3D grid map to map the obstacle information and the tactile display. And the obstacle information is transferred to a blind pedestrian using tactile feedback. Based on the obstacle information a user plans a walking route and controls the guide vehicle. The algorithm for 3D detection of an obstacle distribution and the method of mapping the generated obstacle map and the tactile display device are proposed in this paper. The experiment for the 3D detection of an obstacle distribution using ultrasonic sensors is performed and estimated. The experimental system consisted of ultrasonic sensors and control system. In the experiment, the detection of fixed obstacles on the ground, the moving obstacle, and the detection of down-step are performed. The performance for the 3D detection of an obstacle distribution and space mapping is verified through the experiment.