隨著科技的進步, 人們紀錄生活從文字敘述慢慢變成以影像來記錄, 所以影像資料成長速度非常快速, 因為這個原因, 如何管理影像資料庫及如何從這龐大資料中快速且正確地找出我們所要的影像, 確實成為一個值得我們深入探討研究的課題. 現今影像檢索的方法通常分為兩個步驟: 第一, 先對資料庫中的影像進行分析, 以所謂的特徵值來代表影像. 第二, 將查詢的影像資料以相同的方法去分析取得代表的特徵向量, 對資料庫中的影像特徵值比對, 以獲得相似的影像.
我們提出的方法, 使用影像的邊界形狀用三種不同的特徵值來對影像作取回的動作. 首先我們對查詢影像作邊界偵測, 紀錄所需要使用的特徵值, 在我們比對的方法, 使用到一個比較點集的差異性來作為兩者形狀的差異的方法, 當一個點對應到另一個點找到最小差值時, 下一個點的對應方法有兩種: 第一, 我們在比對第二個點時, 並不排除前一次比對到的點, 所以有可能會多點對到同一個點. 第二, 我們將第一次對到的點再下一個點對應時拿走, 這養子可以保證我們每一個點都對應到不同點, 但是當兩邊點數不相同時, 就會發生有對應不到點的情形發生, 所以得到的懲罰值會變很大.

我們使用三種不同特徵值及兩種不同計算方法, 去做這一個實驗. 現在我們可以正確且快速的去取回所查詢的影像, 接下來改進的方法要去提高飛查詢影像的相似性, 或者結合其他方法來提高搜尋正確性.

Recently, the management of the database becomes more important. And it needs to find a method to search the queried image fast and correctly. Image database are usually implemented in two steps. First, we analyze the images in the database offline using some visual content attributes, and store the features to represent the image. Then, we apply the same analysis to the query image to compare the features to find out the similar or correct image. In the image retrieval method, there are many content attributes are adopted to measure the similarity between two images. In our method, we use the Sobel edge detection and proper threshold to detect the image to find out the image shape. Then, we use Hausdorff distance to compute the dissimilarity. We apply three measures to compute Hausdorff distance. First, we use gray level values to be the distance measure due to simple. Second, we use Euclidean distance. This measure can catch the exactly matching pixel location. The last one, we use the gradient of the image as the distance measure. This supports more information to be the comparison features; it can get the better query results.

[1] F. Idris and S. Panchanathan, Review of Image and Video Indexing Techniques, Journal of Visual Communication and Image Representation, Vol 8, pp. 146-166, 1997
[2] Jose M. Martinez, Overview of the MPEG-7 Stardard(version 5.0)
[3] Philippe Salembier and John R. Smith, MPEG-7 Multimedia Description Schemes, IEEE Transactions on Circuits and Systems for Video Technology, Vol 11, 2001
[4] Sadegh Abbasi and Farzin Mokhtarian, Affine-Similar Shape Retrieval: Application to Multiview 3-D Object Recognition, IEEE Transaction on Image Processing, Vol 10, 2001
[5] Paolo Gastaldo and Rodolfo Zunino, Hausdorff Distance For Target Detection, IEEE International Symposium on, Volume: 5, p.661-664, 2002
[6] Qian Huang, Atul Puri and Zhu Liu, Multimedia Search and Retrieval: New Concepts, System Implementation, and Application, IEEE Transactions on Circuits and Systems for Video Technology, Vol 10, 2000
[7] Elif Albuz, Erturk Kocalar and Ashfaq A. Khokhar, Scalable Color Image Indexing and Retrieval Using Vector Wavelets, IEEE Transactions on knowledge and Data Engineering, Vol 13, 2001
[8] Soo-Chang Pei and Ching-Min Cheng, Extracting Color Features and Dynamic Matching foe Image Data-Base Retrieval, IEEE Transactions on Circuits and Systems for Video Technology, Vol 9, 1999
[9] Byung Cheol Sing, Myung Jun Kim and Jong Beom Ra, A Fast Multi-resolution Feature Maching Algorithm for Exhaustive Search in Large Image Database, IEEE Transactions on Circuits and Systems for Video Technology, Vol 11, 2001
[10] Yoram Gdalyahu, Daphna Weinshall and Michael Werman, Self-Organization in Vision: Stochastic Clustering for Image Segmentation, Perceptual Grouping, and Image Database Organization, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 23, 2001
[11] Simone Santini, Amarnath Gupta and Ramesh Jain, Emergent Semantics through Interaction in Image Databases, IEEE Transactions on knowledge and Data Engineering, Vol 13, 2001
[12] Fabio Dell’Acqua and Paolo Gamba, Simplified Model Analysis and Search for Reliable Shape Retrieval, IEEE Transactions on Circuits and Systems for Video Technology, Vol 8, 1998
[13] Linhui Jia and Leslie Kitchen, Object-Based Image Similarity Computation Using Inductive Learning of Contour-Segment Relations, IEEE Transactions on Image Processing, Vol 9, 2000
[14] Donald A. Adjeroh and M. C. Lee, On Ratio-Based Color Indexing, IEEE Transactions on Image Processing, Vol 10, 2001
[15] Linhui Jia and Leslie Kitchen, Object-Based Image Similarity Computation Using Inductive Learning of Contour-Segment Relations, IEEE Transactions on Image Processing, Vol 9, 2000
[16] Xilin Yi and Octavia I. Camps, Line-Based Recognition Using A Multidimensional Hausdorff Distance, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 21, 1999
[17] Jungwoo Lee and Bradley W. Dickinson, Hierchical Video Indexing and Retrieval for Subband-Coded Video, IEEE Transactions on Circuits and Systems for Technology, Vol 10, 2000
[18] Inseo Han, Il Dong Yun and Sang Uk Lee, Model-based Object Recognition Using the Hausdorff Distance with Explicit Pairing, International conference on Image Processing , Vol 4, 1999