為了更有效、更準(zhǔn)確地進(jìn)行的圖像檢索，在迭代函數(shù)系統(tǒng)收斂速度理論和拼貼定理的基礎(chǔ)上，提出了一種結(jié)合分形編碼進(jìn)行基于內(nèi)容的圖像檢索方法，即把查詢圖像的分形碼應(yīng)用于圖像庫(kù)中的圖像進(jìn)行分形迭代，然后將得到的圖像與查詢圖像進(jìn)行相似匹配，在檢索正確率和檢索速度方面，優(yōu)于實(shí)驗(yàn)中其他方法。實(shí)驗(yàn)證明了該算法的有效性和可行性。
關(guān)鍵詞：分形編碼；迭代函數(shù)系統(tǒng)；基于內(nèi)容的圖像檢索

A New Method of Texture-Based Image Retrieval Combining Fractal Coding Zhang Hong-xia Ma Yan Shi Shu-jian (Department of Computer, Shanghai Normal University, china, 200234) 【Abstract】Based on the theory of the speed of convergence for iterative function system and Collage theorem, a new approach of content-based image retrieval combining fractal coding is proposed to retrieve images effectively in this paper. The fractal codes of the query image is applied and iterated on the images in the database. The iterated image is matched with the query image. The proposed method is better than other methods in accuracy and speed. The experiments show that the proposed method is effective and feasible.
【Key words】 Fractal Coding; Iterative Function System; Content-based Image Retrieval