畢業(yè)論文基于小波變換手指靜脈去噪技術(shù)-免費(fèi)閱讀
【正文】 最后對(duì)處理后的各頻層作靜態(tài)小波逆變換，得到降噪后的圖像。所以，對(duì)于中頻帶來(lái)說(shuō)，我們只保留圖像中的橫向和縱向極值點(diǎn)，而對(duì)于其他點(diǎn)都置為0。隨著分解步數(shù)增大，濾波器也逐步增大。靜態(tài)小波分解分為兩步：1) 給定長(zhǎng)度為N的信號(hào)s，SWT的第一步是將s分成兩個(gè)系數(shù)集：低頻系數(shù)（逼近系數(shù)）和高頻系數(shù)（細(xì)節(jié)系數(shù)），前者通過(guò)s和低通濾波器Lo_D卷積得到，而后者是和高通濾波器Hi_D卷積得到的。以下是以lena圖像為例，利用Mallat算法將圖像進(jìn)行兩級(jí)分解示意圖：圖22圖像兩層小波分解示意圖程序見(jiàn)附錄一。關(guān)于Riesz的具體說(shuō)明如下：若是的Risez基，則存在常數(shù)A，B使得：對(duì)所有雙無(wú)限可平方和序列，即成立。若a, b是連續(xù)變量，稱(chēng)為連續(xù)小波，若a, b 是離散變量，稱(chēng)為離散小波。2 小波變換基本理論上個(gè)世紀(jì)80年代初，Morlet和Arens等人首次提出了“小波”的概念。它是世界上最尖端的具有高精度、高速度的認(rèn)證技術(shù)。關(guān)鍵詞: 靜態(tài)小波變換；圖像去噪；逼近信號(hào)；細(xì)節(jié)信號(hào)Application of SWT Theory for Finger Vein in Image Denoising Abstract: In fingervein recognition, since the acquisition process is affected by fingervein equipment, lighting and other factors, the contrast of the vein image is often low, have much noise, and it brings great inconvenience to the final recognition. So before identifying vein image denoising processing is necessary. This article applied the SWT theory for finger vein in image denoising. Firstly, applied twodimensional static wavelet transform to the original image, then we made use of the different characteristics of each frequency band to design different methods of denoising respectively. Experiments show the method has integrated advantage for image denoising and for enhancing the clarity of the fingervein image.
數(shù)學(xué)相關(guān)推薦
鄂ICP備17016276號(hào)-1