基於單攝影機技術之室內影像定位系統設計 應用於自主式機械人

A design of a single CCD-based indoor localization technology-applied to autonomous mobile robots

賴拓文(To-Wen Lai)

指導教授 ： 王偉彥

國立臺灣師範大學/科技學院/應用電子科技學系/碩士(2012年)

摘要

本論文不同於傳統雙影像設備距離量測方法，採用以平行線距離量測系統(Parallel Lines Distance Measurement System, PLDMS)來實現單一影像設備距離量測，不僅可降低成本且環境架設容易，而且只需知道影像設備的最大像素值、視角及光學距離等環境參數。由於本文之影像設備為固定單一位置，因此採用運算速度較快的背景差值法來提取前景，同時使用低通濾波器進行背景更新以降低背景噪聲，亦使用影像型態學方法來提取完整前景資訊及去除細微雜訊。在機械人室內定位實驗上，我們經由室內的平面地圖規劃，將網路攝影機(Webcam)設備架設於最合適處來監測移物體，再透過所提出單影像設備測距的方法來實現定位。本論文將三台影像設備分別架設於三個走廊轉角處，藉由事先定義的影像設備全域座標，我們可透過單一影像定位獲取移動物體之座標。最後，再透過影像設備間切換機制進而得到完整的全域座標資訊。

關鍵字

影像量測 ； PLDMS ； 室內定位系統

並列摘要

Unlike traditional Binocular vision measurement method, this thesis presents a single-webcam-based measurement method developed from a proposed Parallel Lines Distance Measurement System (PLDMS). PLDMS can create the identical ruler for all measured objects. Not only can the proposed measurement method reduce production cost, but also the experimental environment is easy to set up because only three parameters need to decide, the maximum pixel, perspective, and optical distance. Because the locations of webcams are fixed, we use the simple background subtraction method to extract the prospects to improve the problem of computational burden. Furthermore, we use the low-pass filter and on-line background update method to reduce background noise, and adopt the image morphology to complete prospect information and to remove the slight noise. In our indoor experiments, webcams are located several places on where we can clearly monitor the move of a robot in the fifth floor of the Science and Technology building of Nation Taiwan Normal University. Finally, through the switching mechanism and the predefined coordinate system, we can get the location of the robot when it is moving.

並列關鍵字

Image-based distance measuring system ； PLDMS ； Indoor localization system

參考文獻

[1] J. B. Kim and H. S. Jun, “Vision-Based Location Positioning using Augmented Reality for Indoor Navigation,” IEEE Transactions on Consumer Electronic, vol. 54, no. 3, Aug. 2008.

[2] F. Serratosa and A. Sanfeliu, “Vision-Based Robot Positioning by an Exact Distance Between Histograms,” The 18th International Conference on Pattem Recognition (ICPR’09).

[5] C. Shengli and Z. Huiqing, “Research on Indoor Position Algorithm Based on Image Sequence,” 2010 International Conference on Computer Design And Appliations (ICCDA 2010).

[6] M. Cypriani, F. Lassabe, P. Canalda, and F. Spies, “Wi-Fi-Based Indoor Positioning: Basic Techniques, Hybrid Algorithms and Open Software Platform,” 2010 International Conference On Indoor Positioning And Indoor Navigation (IPIN), pp.15-17 Sept. 2010, Zurich, SWITZERLAND.

[8] P. Bahl and V. N. Padmanabhan, “RADAR: An In-bulding RF-based User Location and Tracking System,” IEEE Computer and Communications Conference, vol. 2, 2000, pp. 775-784.