以智慧型手機做戶外街景之擴增實境導覽

Abstract

人們來到一個戶外旅遊場所時，常會有迷路或是不知道如何到達想去地點或建築物的情形。對此，本研究提出一個利用電腦視覺技術，應用於戶外旅遊場所的擴增實境式導覽系統。所提系統事先建立一導覽地圖，提供遊客旅遊資訊，並有兩個主要功能：旅遊導覽及建築物資訊介紹，皆是建立在擴增實境的技術上。此系統可以指引遊客到事先選取的建築物，以及將建築物資訊擴增到使用者所持智慧型手機的螢幕上。 為了完成上述功能，首先要建立一環境地圖，包含旅遊路線區域平面圖及沿線建築物資訊；並以智慧手機的相機拍攝整段路線上建築物的影像，並記錄其對應位置；最後將所有資料儲存到一個資料庫，供導覽階段使用。 接下來，本研究提出一個以加速穩健特徵(SURFs)做影像辨識的演算法，來對使用者進行的方法。該方法首先要求使用者對鄰近建築物拍攝影像，接著利用加速穩健特徵，將建築物影像與事先建立的影像資料庫做比對，進行建築物辨識。最後，根據使用者拍攝的影像與比對結果之間的透視關係，完成使用者定位，並獲得使用者位置及方向參數。 此外，本研究提出一擴增實境式導覽與建築物資訊介紹的方法。根據使用者的位置及方向參數，所提系統會利用戴克斯特拉(Dijkstra)演算法，規劃一條從使用者位置到一事先選取的目的地之間的最短路徑，並在智慧手機螢幕中的沿路影像上貼上一擴增實境式箭頭，來指引使用者該往哪裡走。此外，也在智慧手機螢幕上擴增眼前所見建築物之名稱，以及建築物與使用者之間的距離。 最後，本研究以良好的實驗結果證明所提系統確實可行。

When people go to an outdoor tour area, they might get lost or have no idea about how to reach desired locations or buildings. In this study, an augmented reality (AR)-based tour guidance system for use in outdoor tour areas by computer vision techniques is proposed. With the proposed system, a tour guidance map can be constructed in advance to provide tour guidance information to a tourist. The system provides two main functions: tour guidance and building-information introduction, both based on AR techniques. The system can guide a touring user to the location of a pre-selected building and augment the information of the building on the screen of a smart-phone held by the user. To implement a system with the above-mentioned capabilities, at first an environment map is generated in a learning phase, which includes the top-view map of a selected tour path area and the information about the along-path buildings. In addition, the images of the buildings taken by the smart-phone camera and the corresponding building locations along the whole tour path are also learned. All the learned data are saved into a database for use in the navigation phase. Next, a method for user localization is proposed for use by the system. The method is based on image recognition using speeded up robust features (SURFs). At first, the user takes an image of a nearby building. Secondly, recognition of the building image is conducted against the pre-constructed image database by an SURF matching algorithm. Finally, user localization is accomplished by the system to obtain the user’s position and orientation parameters according to the perspective relationship between the image taken by the user as well as the matching result. Furthermore, a method for AR-based guidance and building-information introduction is proposed. Based on the user’s position and orientation parameters, the system can plan a shortest path from the user’s location to a pre-selected destination by use of the Dijkstra algorithm. Accordingly, an AR-based arrow is created and augmented on the acquired along-path building image on the smart-phone screen to guide the user where to go. Also augmented on the smart-phone screen is the information of the building name as well as the distance between the user and the building appearing in the acquired image which is computed according to the user localization result. Good experimental results are also presented to show the feasibility of the proposed methods and system for real applications.