• International Journal of Naval Architecture and Ocean Engineering
• /
• 제11권1호
• /
• pp.202-210
• /
• 2019

The obstacles modeling is a fundamental and significant issue for path planning and automatic navigation of Unmanned Surface Vehicle (USV). In this study, we propose a novel obstacles modeling method based on high resolution satellite images. It involves two main steps: extraction of obstacle features and construction of convex hulls. To extract the obstacle features, a series of operations such as sea-land segmentation, obstacles details enhancement, and morphological transformations are applied. Furthermore, an efficient algorithm is proposed to mask the obstacles into convex hulls, which mainly includes the cluster analysis of obstacles area and the determination rules of edge points. Experimental results demonstrate that the models achieved by the proposed method and the manual have high similarity. As an application, the model is used to find the optimal path for USV. The study shows that the obstacles modeling method is feasible, and it can be applied to USV path planning.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.195-206
• /
• 2022

Crack detection is essential for inspection of existing structures and crack segmentation based on deep learning is a significant solution. However, datasets are usually one of the key issues. When building a new dataset for deep learning, laborious and time-consuming annotation of a large number of crack images is an obstacle. The aim of this study is to develop an approach that can automatically select a small portion of the most informative crack images from a large pool in order to annotate them, not to label all crack images. An active learning method with difficulty learning mechanism for crack segmentation tasks is proposed. Experiments are carried out on a crack image dataset of a steel box girder, which contains 500 images of 320×320 size for training, 100 for validation, and 190 for testing. In active learning experiments, the 500 images for training are acted as unlabeled image. The acquisition function in our method is compared with traditional acquisition functions, i.e., Query-By-Committee (QBC), Entropy, and Core-set. Further, comparisons are made on four common segmentation networks: U-Net, DeepLabV3, Feature Pyramid Network (FPN), and PSPNet. The results show that when training occurs with 200 (40%) of the most informative crack images that are selected by our method, the four segmentation networks can achieve 92%-95% of the obtained performance when training takes place with 500 (100%) crack images. The acquisition function in our method shows more accurate measurements of informativeness for unlabeled crack images compared to the four traditional acquisition functions at most active learning stages. Our method can select the most informative images for annotation from many unlabeled crack images automatically and accurately. Additionally, the dataset built after selecting 40% of all crack images can support crack segmentation networks that perform more than 92% when all the images are used.

• 전자공학회논문지SC
• /
• 제47권6호
• /
• pp.86-96
• /
• 2010

본 논문에서는 U-시차맵과 조감도를 이용한 스테레오 비전 기반의 장애물체 검출 및 차량 검증 방법을 제안한다. 먼저 최대 빈도 값을 이용하여 V-시차맵 상에서 도로 특징 정보를 추출하고, 추출된 도로 정보를 이용하여 대략적인 도로상의 장애물체 영역을 추출한다. 좀 더 정확한 장애물체 영역 추출을 위하여 U-시차맵을 생성하는데, 이때 시차값과 카메라 파라미터를 이용하여 계산된 문턱치를 이용하여 높이 제한된 U-시차맵을 생성함으로써, 일정한 높이의 장애물체만을 검출 할 수 있다. 그러나 검출된 장애물체 영역 내에는 여전히 다수의 장애물체와 배경이 존재하므로, 세그먼테이션 과정을 수행한다. 전 단계에서 추출된 장애물체 영역을 카메라 모델링과 파라미터를 이용하여 조감도 맵핑을 수행한다. 조감도는 시차맵과 카메라 정보를 기반으로 계산된 장애물체들의 위치를 평면상에 표시함으로써 장애물체들을 좀 더 쉽게 분리할 수 있다. 마지막으로 각각 분리된 장애물체들 별로 차량 특징 기반의 차량 검증 과정을 수행한다. 도로 접점 여부, 일정한 수평크기, 가로 세로 비율 및 텍스쳐 정보를 이용하여 최종적으로 도로상의 차량만을 검출한다. 그리고 실제 도로에서 획득한 영상에 제안한 알고리즘을 적용함으로써 장애물체 검출 및 차량 검증 성능을 검증한다.

• 센서학회지
• /
• 제33권4호
• /
• pp.187-195
• /
• 2024

Light detection and ranging (LiDAR), a widely used sensor in mobile robots and autonomous vehicles, has its most important function as measuring the range of objects in three-dimensional space and generating point clouds. These point clouds consist of the coordinates of each reflection point and can be used for various tasks, such as obstacle detection and environment recognition. However, several processing steps are required, such as three-dimensional modeling, mesh generation, and rendering. Efficient data processing is crucial because LiDAR provides a large number of real-time measurements with high sampling frequencies. Despite the rapid development of controller computational power, simplifying the computational algorithm is still necessary. This paper presents a method for estimating the presence of curbs, humps, and ground tilt using range measurements from a single horizontal or vertical scan instead of point clouds. These features can be obtained by data segmentation based on linearization. The effectiveness of the proposed algorithm was verified by experiments in various environments.

• 한국게임학회 논문지
• /
• 제17권6호
• /
• pp.139-152
• /
• 2017

본 논문에서는 구면 영상에서 영역 분할 정보를 사용하여 바닥 영역을 검출하는 방법을 제시한다. 평면 영상에서의 Watershed 영역 분할 방법을 수정하여 구면 영상의 영역 분할에 적용할 수 있도록 하였다. 영역들을 분할한 뒤 가정된 바닥 영역 픽셀의 색상과 질감을 그 외의 영역들과 비교하여 바닥 영역을 검출한다. 구면 파노라마 영상에서는 구면 왜곡으로 인하여 평면에서의 바닥 검출 방법을 그대로 적용할 수 없다. 구면 왜곡을 고려한 바닥 영역 검출을 위하여 바닥 영역의 외곽선을 검출하는 알고리즘을 설계하였다. 실험에서 지상물이 없는 경우와 있는 경우의 모두에서 적절하게 바닥 영역을 검출할 수 있는 결과를 보였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.163-166
• /
• 2000

Real-time traffic detection scheme based on Computer Vision is capable of efficient traffic control using automatically computed traffic information and obstacle detection in moving automobiles. Traffic information is extracted by segmenting vehicle region from road images, in traffic detection system. In this paper, we propose the advanced segmentation of vehicle from road images using multiple local region information. Because multiple local region overlapped in the same lane is processed sequentially from small, the traffic detection error can be corrected.

• 한국멀티미디어학회논문지
• /
• 제19권6호
• /
• pp.1003-1013
• /
• 2016

This paper proposes a framework of superpixel-based vehicle detection method using plane normal vector in disparity space. We utilize two common factors for detecting vehicles: Hypothesis Generation (HG) and Hypothesis Verification (HV). At the stage of HG, we set the regions of interest (ROI) by estimating the lane, and track them to reduce computational cost of the overall processes. The image is then divided into compact superpixels, each of which is viewed as a plane composed of the normal vector in disparity space. After that, the representative normal vector is computed at a superpixel-level, which alleviates the well-known problems of conventional color-based and depth-based approaches. Based on the assumption that the central-bottom of the input image is always on the navigable region, the road and obstacle candidates are simultaneously extracted by the plane normal vectors obtained from K-means algorithm. At the stage of HV, the separated obstacle candidates are verified by employing HOG and SVM as for a feature and classifying function, respectively. To achieve this, we trained SVM classifier by HOG features of KITTI training dataset. The experimental results demonstrate that the proposed vehicle detection system outperforms the conventional HOG-based methods qualitatively and quantitatively.

• 대한기계학회논문집A
• /
• 제21권10호
• /
• pp.1551-1560
• /
• 1997

This study is devoted to the detection of the 3-dimensional point obstacles on the plane by using accumulated scan line images. The proposed accumulating only one scan line allow to process image at real time. And the change of motion of the feature in image is small because of the short time between image frames, so it does not take much time to track features. To obtain recursive optimal obstacles position and robot motion along to the motion of camera, Kalman filter algorithm is used. After using Kalman filter in case of the fixed environment, 3-dimensional obstacles point map is obtained. The position and motion of moving obstacles can also be obtained by pre-segmentation. Finally, to solve the stereo ambiguity problem from multiple matches, the camera motion is actively used to discard mis-matched features. To get relative distance of obstacles from camera, parallel stereo camera setup is used. In order to evaluate the proposed algorithm, experiments are carried out by a small test vehicle.

• 한국건설관리학회논문집
• /
• 제5권5호
• /
• pp.151-162
• /
• 2004

Current methods for construction site modeling employ large, expensive laser range scanners that produce dense range point clouds of a scene from different perspectives. Days of skilled interpretation and of automatic segmentation may be required to convert the clouds to a finished CAD model. The dynamic nature of the construction environment requires that a real-time local area modeling system be capable of handling a rapidly changing and uncertain work environment. However, in practice, large, simple, and reasonably accurate embodying volumes are adequate feedback to an operator who, for instance, is attempting to place materials in the midst of obstacles with an occluded view. For real-time obstacle avoidance and automated equipment control functions, such volumes also facilitate computational tractability. In this research, a human operator's ability to quickly evaluate and associate objects in a scene is exploited. The operator directs a laser range finder mounted on a pan and tilt unit to collect range points on objects throughout the workspace. These groups of points form sparse range point clouds. These sparse clouds are then used to create geometric primitives for visualization and modeling purposes. Experimental results indicate that these models can be created rapidly and with sufficient accuracy for automated obstacle avoidance and equipment control functions.

• ETRI Journal
• /
• 제40권6호
• /
• pp.699-713
• /
• 2018

Orthogonal frequency division multiplexing (OFDM) has been the overwhelmingly prevalent choice for high-data-rate systems due to its superior advantages compared with other modulation techniques. In contrast, a high peak-to-average-power ratio (PAPR) is considered the fundamental obstacle in OFDM systems since it drives the system to suffer from in-band distortion and out-of-band radiation. The partial transmit sequence (PTS) technique is viewed as one of several strategies that have been suggested to diminish the high PAPR trend. The PTS relies upon dividing an input data sequence into a number of subblocks. Hence, three common types of the subblock segmentation methods have been adopted - interleaving (IL-PTS), adjacent (Ad-PTS), and pseudorandom (PR-PTS). In this study, a new type of subblock division scheme is proposed to improve the PAPR reduction capacity with a low computational complexity. The results indicate that the proposed scheme can enhance the PAPR reduction performance better than the IL-PTS and Ad-PTS schemes. Additionally, the computational complexity of the proposed scheme is lower than that of the PR-PTS and Ad-PTS schemes.