• 제목/요약/키워드: Automatic Data Extraction

검색결과 311건 처리시간 0.03초

Quadtree와 영역확장법에 의한 LiDAR 데이터의 지면점 추출 (Extraction of Ground Points from LiDAR Data using Quadtree and Region Growing Method)

  • 배대섭;김진남;조기성
    • 대한공간정보학회지
    • /
    • 제19권3호
    • /
    • pp.41-47
    • /
    • 2011
  • 원시 LiDAR 데이터는 벡터 구조이기 때문에 직접 활용 시 처리과정이 복잡해지지만, LiDAR 데이터를 필터링을 통해 정규 가상 격자 형태로 변환하면 데이터 용량이 감소되고 처리 속도가 빠르기 때문에 저가의 장비에서도 처리가 가능하다. 특히 Quadtree와 같은 영상 압축 처리 기법을 적용할 경우, 평활화를 통하여 비지면 요소인 자동차, 수목등이 제거되어 모델링에 유리하다는 장점이 있다. 따라서 본 연구에서는 대용량의 LiDAR 데이터로부터 Quadtree와 영역확장법을 활용하여 지면점을 자동 추출할 수 있는 알고리즘을 제시하였으며, 오차분류기법을 활용하여 정확도를 분석하였다. 그 결과, 지면점 분류 정확도는 98%이상으로 나타나, 지면점 추출에 유리함을 알 수 있었다. 또한 Quadtree와 영역확장법을 활용시 자동차, 수목등의 비지면 요소들을 효과적으로 제거할 수 있었다.

Oil Pipeline Weld Defect Identification System Based on Convolutional Neural Network

  • Shang, Jiaze;An, Weipeng;Liu, Yu;Han, Bang;Guo, Yaodan
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제14권3호
    • /
    • pp.1086-1103
    • /
    • 2020
  • The automatic identification and classification of image-based weld defects is a difficult task due to the complex texture of the X-ray images of the weld defect. Several depth learning methods for automatically identifying welds were proposed and tested. In this work, four different depth convolutional neural networks were evaluated and compared on the 1631 image set. The concavity, undercut, bar defects, circular defects, unfused defects and incomplete penetration in the weld image 6 different types of defects are classified. Another contribution of this paper is to train a CNN model "RayNet" for the dataset from scratch. In the experiment part, the parameters of convolution operation are compared and analyzed, in which the experimental part performs a comparative analysis of various parameters in the convolution operation, compares the size of the input image, gives the classification results for each defect, and finally shows the partial feature map during feature extraction with the classification accuracy reaching 96.5%, which is 6.6% higher than the classification accuracy of other existing fine-tuned models, and even improves the classification accuracy compared with the traditional image processing methods, and also proves that the model trained from scratch also has a good performance on small-scale data sets. Our proposed method can assist the evaluators in classifying pipeline welding defects.

2차원 칼라 얼굴 영상에서 반복적인 PCA 재구성을 이용한 자동적인 잡음 제거 (Automatic Denoising of 2D Color Face Images Using Recursive PCA Reconstruction)

  • 박현;문영식
    • 전자공학회논문지CI
    • /
    • 제43권2호
    • /
    • pp.63-71
    • /
    • 2006
  • 칼라 영상의 잡음 제거 및 복원은 컴퓨터 비젼 및 영상 처리 분야에서 점점 더 많은 연구가 되어지고 있는 분야이다. 칼라얼굴 영상에서의 잡음 제거 및 복원은 색상들 간의 미묘한 상호작용뿐만 아니라 얼굴의 구조학적 특징 때문에 일반적인 영상의 처리보다 더욱 어렵다. 본 논문은 벡터기반의 영상 필터들을 이용하여 제거하기 어려운 칼라 얼굴 영상의 복합 잡음을 제거 하기 위해 PCA 재구성 기반의 잡음 제거 방법을 제안한다. 제안된 방법은 PCA를 이용한 정준 고유얼굴 공간의 학습단계, 동적 외양 모델을 이용한 자동적인 얼굴 특징 추출 단계, Bilateral 필터를 이용한 복원된 칼라 영상의 재조명(Relighting) 단계, 학습 데이터들의 분산 값들을 이용한 잡음 영역 추출 단계, 입력 영상의 부분 정보를 이용한 재구성과 이를 원본 영상과 합성하여 잡음이 제거된 영상을 생성하는 단계 등 총 5 단계로 구성된다. 실험결과는 제안된 방법이 입력 얼굴 영상들의 구조적 특징들은 잘 유지하면서 복합적인 칼라 잡음 등을 효과적으로 제거하는 것을 보인다.

자동 특징 추출기법에 의한 최소의 주식예측 특징선택 (Minimized Stock Forecasting Features Selection by Automatic Feature Extraction Method)

  • 이상홍;임준식
    • 한국지능시스템학회논문지
    • /
    • 제19권2호
    • /
    • pp.206-211
    • /
    • 2009
  • 본 논문은 가중 퍼지소속함수 기반 신경망(Neural Network with Weighted Fuzzy Membership Functions, NEWFM)기반의 자동 특징 추출기법을 사용하여 1일 후의 주식 예측을 하는 방안을 제안하고 있다. 비중복면적 분산측정 법에 의해 중요도가 가장 낮은 특징입력을 자동적으로 하나씩 제거하면서 최소의 특징입력을 선택하였다. 특징입력으로써 CPP$_{n,m}$(Current Price Position of the day n)과 최근 32일간의 CPP$_{n,m}$을 웨이블릿 변환한 38개의 계수들 중 비중복면적 분산측정법에 의해서 자동적으로 추출된 2개의 계수가 사용되었다 제안된 방법으로 1989년부터 1998년까지의 실험군을 사용한 결과로써 60.93%의 예측율을 나타내었다.

AUTOMATIC IDENTIFICATION OF ROOF TYPES AND ROOF MODELING USING LIDAR

  • Kim, Heung-Sik;Chang, Hwi-Jeong;Cho, Woo-Sug
    • 대한원격탐사학회:학술대회논문집
    • /
    • 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
    • /
    • pp.83-86
    • /
    • 2005
  • This paper presents a method for point-based 3D building reconstruction using LiDAR data and digital map. The proposed method consists of three processes: extraction of building roof points, identification of roof types, and 3D building reconstruction. After extracting points inside the polygon of building, the ground surface, wall and tree points among the extracted points are removed through the filtering process. The filtered points are then fitted into the flat plane using ODR(Orthogonal Distance Regression). If the fitting error is within the predefined threshold, the surface is classified as a flat roof. Otherwise, the surface is fitted and classified into a gable or arch roof through RMSE analysis. Based on the roof types identified in automated fashion, the 3D building reconstruction is performed. Experimental results showed that the proposed method classified successfully three different types of roof and that the fusion of LiDAR data and digital map could be a feasible method of modelling 3D building reconstruction.

  • PDF

Discriminative Models for Automatic Acquisition of Translation Equivalences

  • Zhang, Chun-Xiang;Li, Sheng;Zhao, Tie-Jun
    • International Journal of Control, Automation, and Systems
    • /
    • 제5권1호
    • /
    • pp.99-103
    • /
    • 2007
  • Translation equivalence is very important for bilingual lexicography, machine translation system and cross-lingual information retrieval. Extraction of equivalences from bilingual sentence pairs belongs to data mining problem. In this paper, discriminative learning methods are employed to filter translation equivalences. Discriminative features including translation literality, phrase alignment probability, and phrase length ratio are used to evaluate equivalences. 1000 equivalences randomly selected are filtered and then evaluated. Experimental results indicate that its precision is 87.8% and recall is 89.8% for support vector machine.

온톨로지를 이용한 이미지의 고수준 의미 정보 자동 추출 기법 (Full-automatic high-level concept extraction for image using domain ontologies)

  • 박경욱;이동호
    • 한국정보과학회:학술대회논문집
    • /
    • 한국정보과학회 2005년도 가을 학술발표논문집 Vol.32 No.2 (2)
    • /
    • pp.88-90
    • /
    • 2005
  • 최근 인터넷의 급속한 성장은 이미지와 같은 멀티미디어 정보의 급격한 증가를 가져왔다. 따라서 사용자로 하여금 원하는 이미지를 검색하는데 있어서 좀 더 효율적이고 정확한 검색 방법의 필요성이 대두되어 왔다. 일반적으로 이미지 검색 방법에는 키워드 기반 방식과 내용 기반 방식이 존재한다. 그러나 위 두 방법은 지금의 대용량 이미지 데이터베이스 검색에 있어서 여러 문제점들을 가지고 있다. 특히, 키워드 기반 방식을 보완하기 위해서 제안되어진 내용 기반 방식의 경우, 사람이 인식할 수 있는 의미 정보가 아닌 시각 정보만을 이용하기 때문에 시맨틱 갭(semantic gap) 문제가 발생하게 된다. 본 논문에서는 이미지 객체의 시각 정보들에 대한 중간 의미값으로 구성된 시각 정보 온톨로지와 동물에 대한 분류 정보를 표현하고 있는 동물 온톨로지를 구축하고, 이를 이용하여 이미지로부터 .고수준의 의미 정보를 완전 자동으로 추출하는 효율적인 방법을 제안한다.

  • PDF

Spatio-Temporal Analysis of Trajectory for Pedestrian Activity Recognition

  • Kim, Young-Nam;Park, Jin-Hee;Kim, Moon-Hyun
    • Journal of Electrical Engineering and Technology
    • /
    • 제13권2호
    • /
    • pp.961-968
    • /
    • 2018
  • Recently, researches on automatic recognition of human activities have been actively carried out with the emergence of various intelligent systems. Since a large amount of visual data can be secured through Closed Circuit Television, it is required to recognize human behavior in a dynamic situation rather than a static situation. In this paper, we propose new intelligent human activity recognition model using the trajectory information extracted from the video sequence. The proposed model consists of three steps: segmentation and partitioning of trajectory step, feature extraction step, and behavioral learning step. First, the entire trajectory is fuzzy partitioned according to the motion characteristics, and then temporal features and spatial features are extracted. Using the extracted features, four pedestrian behaviors were modeled by decision tree learning algorithm and performance evaluation was performed. The experiments in this paper were conducted using Caviar data sets. Experimental results show that trajectory provides good activity recognition accuracy by extracting instantaneous property and distinctive regional property.

Research of fast point cloud registration method in construction error analysis of hull blocks

  • Wang, Ji;Huo, Shilin;Liu, Yujun;Li, Rui;Liu, Zhongchi
    • International Journal of Naval Architecture and Ocean Engineering
    • /
    • 제12권1호
    • /
    • pp.605-616
    • /
    • 2020
  • The construction quality control of hull blocks is of great significance for shipbuilding. The total station device is predominantly employed in traditional applications, but suffers from long measurement time, high labor intensity and scarcity of data points. In this paper, the Terrestrial Laser Scanning (TLS) device is utilized to obtain an efficient and accurate comprehensive construction information of hull blocks. To address the registration problem which is the most important issue in comparing the measurement point cloud and the design model, an automatic registration approach is presented. Furthermore, to compare the data acquired by TLS device and sparse point sets obtained by total station device, a method for key point extraction is introduced. Experimental results indicate that the proposed approach is fast and accurate, and that applying TLS to control the construction quality of hull blocks is reliable and feasible.

Automatic Linkage Method Between Email and Block Structure to Store Construction Project Documents in The Blockchain

  • Kim, Eu Wang;Park, Min Seo;Kim, Jong Inn;Wei, Ameng;Kim, Kyoungmin;Kim, Kyong Ju
    • 국제학술발표논문집
    • /
    • The 9th International Conference on Construction Engineering and Project Management
    • /
    • pp.886-892
    • /
    • 2022
  • In construction projects, it is common to exchange documents using email because of convenience. In this study, a method extracting and organizing block information automatically based on email was developed. This method is composed of document exchange and archiving processes, which are difficult to manage and vulnerable to loss. Therefore, this study aims to develop a solution that can automatically link email and block information. The block data components are designed to derive from email exchange and user-additional input information. Also, automatically generating blocks process including extraction and conversion of information was proposed. This solution can lead to promote the convenience of project document management in terms of identifying the document flow and preventing loss of information.

  • PDF