針對數位學習的圖表/影像檢索

Graph/Image Retrieval for e-Learning Content

黃盈文(Ying-Wen Huang)

指導教授 ： 陳淑媛

元智大學/資訊學院/資訊工程學系/碩士(2005年)

摘要

圖例檢索的目的是為了提供手邊圖例相關資訊的檢索。為了達成此一目的，從資料庫中正確地檢索出相關圖例是必要的。本論文所處理的圖例可以分為兩大類：圖形和影像。所提圖例檢索法分為兩個階段進行：圖例萃取與圖例檢索。在萃取的階段，首先將包含圖例的彩色圖像由RGB轉換為YIQ的灰階圖像，在此Y就是灰階值。接著針對包含圖例之影像進行前景與背景分離，以提供後續圖例萃取與檢索所需要的圖例特徵。圖例可能不會佔滿整張圖像，所以需先在整張圖像中找出只包含圖例的範圍，如此不但可以減少處理的時間同時也增加檢索的正確率。在檢索的階段，是利用萃取圖例的特徵來進行比對的工作，我們使用的特徵包含了圖例長寬比、圖例組合數、具有空間性的前景點數、輪廓長度、及灰階值直方圖。最後，針對檢索圖例類別，採取不同比對函數，以計算檢索圖例和資料庫中的每一張圖例之相似度，以達成正確檢索圖例並獲得圖例相關資訊的目的。最後，經由眾多實驗證實所提方法確實有效且可行。

關鍵字

直方圖比對

並列摘要

The goal of legend retrieval is to provide related information retrieval for the handed legend. Hence, it is necessary to retrieve the correct legend from database for the query legend. The processed legends can be properly divided into two categories: graph and image. The proposed method consists two phases: legend extract and legend retrieval. In the extraction stage, the legend images are first converted from RGB into YIQ color spaces to get Y values as gray-level images. The foreground is then separated from background for the legend image so as to get the characteristics of legend for later legend extraction and retrieval. Since each legend may not occupy the whole legend image, the region enclosing the legend only should be detected first to reduce processing time as well as increase retrieval accuracy. In the retrieval stage, features are extracted for each legend to proceed legend matching. The features used in our method include aspect ratio, number of legend components and spatial histograms of pixel number, border length and gray level. Finally, the strategy of type-based matching is adopted to evaluate the similarity between the query legend and each database legend using different similarity measures according to the type of the query legend. In this way, the correct legend can be retrieved which in turn facilitates the information retrieval for the legend.

並列關鍵字

legend retrieval ； graph retrieval ； image retrieval ； content-based image retrieval ； legend extraction ； histogram-based matching ； e-Learning

參考文獻

[1] I. Bartolini, “Warrp: accurate retrieval of shapes using phase of fourier descriptors and time warping distance,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 27, no. 1, pp. 142-147, 2005.

[2] A. Bishnu, B.B. Bhattacharya, M.K. Kundu, C.A. Murthy and T. Acharya, “Euler vector for search and retrieval of gray-tone image,” IEEE Trans. Systems, Man and Cybernetics, vol. 35, no. 4, pp. 801-812, 2005.

[4] A. Chalechale, G. Naghdy and A. Mertins, “Sketch-based image matching using angular partitioning,” IEEE Trans. Systems, Man and Cybernetics, vol. 35, no. 1, pp. 28-41, 2005.

[5] H. Daschiel and M. Datcu, “Information mining in remote sensing image archives: system evaluation,” IEEE Trans. Geoscience and Remote Sensing, vol. 43, no. 1, pp. 188-199, 2005.

[7] L.A. Fletcher and R. Kasturi, “A robust algorithm for text string separation from mixed text/graphics images,” Pattern Analysis and Machine Intelligence, vol. 10, no. 6, pp. 910-918, 1998.