透過您的圖書館登入
IP:18.220.13.70
  • 期刊

單視角影像滅點偵測與三維建物模型重建

Vanishing Points Detection and 3D Building Models Reconstruction from a Single Perspective View Imagery

摘要


本研究藉由單一視角的平面影像中,自動化地偵測三個彼此正交軸向的滅點,並針對平面建物與三種常見的曲面結構:橢圓柱型、旋轉曲面、自由曲面,進行不同策略性的三維建物重建。整體流程自具有透視投影的照片或是圖畫中萃取出直線特徵,計算滅點與估算特徵點與其對應之特徵底點的位置。接著利用滅點幾何計算特徵點三維坐標,並迭代校正以降低誤差,最後重建曲面建物參數。使用之影像來源包含電腦模擬、視訊擷取、非量測型相機與真實畫作等。建置成果與現地量測等資料進行量化驗証,同時也針對視角條件進行了誤差分析。

並列摘要


This research develops effective algorithms for three mutually orthogonal vanishing points calculation and 3D building models reconstruction from a single perspective view image. In addition, three types of non-planar structures including elliptic cylinder shapes, surfaces of revolution and free-form surface are generated with the developed methods. The proposed scheme begins with line segments detection and filtering from photographs or paintings with fine perspective geometry. Vanishing points, feature points and their corresponding base point were then calculated and minimized random and system errors after iterative fine-tuning algorithms. 3D coordinates of feature points were computed based on vanishing points using single-view metrology. For curved segments, they are extracted separately for particular reconstruction methods. Finally, regular and curved models are then merged based on their shared feature points. The output imagery of a computer-simulated model, video frame cut, consumer camera and real paintings are used in this study to test the performance of the developed algorithms. Quantitative evaluations of the results compared with ground-based surveying and visualized comparison with raw images indicate that, the developed algorithms can successfully extract 3D information and reconstruct 3D models of specific non-planar structures. The accuracy of the proposed methods is related to the viewing angles due to the vanishing point geometry. A validation of viewing angle tolerance is also provided by given random error during vanishing point calculation.

被引用紀錄


Chen, Y. C. (2009). 二維影像對三維影像轉換系統 [master's thesis, National Taipei University of Technology]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0006-2407200918315300
吳姿璇(2015)。整合RGB-D感測器與單眼數位相機的室內環境點雲模型重建〔碩士論文,國立中央大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0031-0412201512094204

延伸閱讀