基于子波變換的紋理圖像分類
TEXTURE CLASSIFICATION BY WAVELET TRANSFORM
-
摘要: 本文用子波變換的方法描述了紋理圖像多尺度、多方向的特性,提出了適合于紋理圖像分類的新的子波特征。通過對(duì)其穩(wěn)定性和視覺特性的詳細(xì)分析,指出此特征優(yōu)于傳統(tǒng)的能量特征。文章最后結(jié)合九類自然紋理圖像,分別基于標(biāo)準(zhǔn)子波特征、子波包特征用BP神經(jīng)網(wǎng)絡(luò)進(jìn)行了分類識(shí)別。實(shí)驗(yàn)結(jié)果表明,在無噪聲情況下,對(duì)自然紋理圖像可無誤差分類;在有噪聲情況下,正確分類識(shí)別率高,表現(xiàn)出強(qiáng)的穩(wěn)定性。
-
關(guān)鍵詞:
- 子波變換; 紋理分類; 特征選擇; BP神經(jīng)網(wǎng)絡(luò)
Abstract: This paper describes the characterization of texture properties at multiple scales and orientations using the wavelet transform, and introduces a new wavelet feature suitable for textured image classification. It is pointed out that the new feature is superior to conventional energy measurement by analyzing its stability and its visual proterty in detail. Finally, nine kinds of natural images are classified successfully based on wavelet feature using BP neural network. The results demonstrate natural textured images can be classified without error and done at higher correct classification rate under white noise. -
Coggins J M, Jain A K. A spatial filtering approach to texture analysis[J].Pattern Recognition Lett.1985, 3:195-203[2]Chang T, Kuo J. Texture analysis and classfication with tree-structured wavelet transform. IEEE Trans. on Image Processing, 1993, IP-2(4): 429-441.[3]Mallat S. Multifrequency channel decomposition of images and wavelets models. IEEE Tans. on ASSP, 1989, ASSP-37(12): 429-441.[4]余越.子波變換理論及其在信號(hào)處理中的應(yīng)用研究:[博士論文]. 北京:北京理工大學(xué),1996.[5]徐朝倫.基于子波變換和模糊數(shù)學(xué)的圖像分割研究:[博士論文]. 北京: 北京理工大學(xué),1998.[6]張靜遠(yuǎn),等.基于小波神經(jīng)網(wǎng)絡(luò)的聲納信號(hào)特征提取與分類.神經(jīng)網(wǎng)絡(luò)理論與應(yīng)用研究96,1996,460-463. -
計(jì)量
- 文章訪問數(shù): 2435
- HTML全文瀏覽量: 90
- PDF下載量: 447
- 被引次數(shù): 0
下載:
