matlab畢業(yè)設(shè)計外文翻譯--復(fù)雜脊波圖像去噪-資料下載頁
2025-05-11 12:58本頁面
【導讀】作者：G.Y.ChenandB.Kegl刊名：PatternRecognition;出版日期：2020. 處理，計算機圖形，IC和模式識別，僅舉幾例。這是因為一個小波變換能結(jié)合的能量,在一小部分的大型系數(shù)和。大多數(shù)的小波系數(shù)中非常小,這樣他們可以設(shè)置為零。這個閾值的小波系數(shù)是?？梢宰龅降闹挥屑毠?jié)的小波分解子帶。我們有一些低頻波子帶不能碰觸，讓。眾所周知，Donoho提出的方法的優(yōu)勢是光滑和自適應(yīng)。因此，他們提出對這些產(chǎn)出去噪通過平均抑制所有循環(huán)信號。結(jié)果證實單目標識別小波消噪優(yōu)于沒有目標識別的情況。果表現(xiàn)出的優(yōu)勢超于了傳統(tǒng)的一對一小波消燥。Chen和Bui擴展這個相鄰小。他們聲稱對于某些標準測試信號和真實圖像相鄰的多。陳等人提出一種圖像去噪是考慮方形相。研究脊波變換的數(shù)多年來打破了小波變換的局限性。最近脊波已成功應(yīng)用于圖像去噪。這種近似二元樹性能的復(fù)雜變性小波和。實驗結(jié)果表明，采用二元。離散脊波變換提供接近理想的稀松代表光滑的物體邊緣。眾所周知，普通的離散小波變換在變換期間是不移位和不轉(zhuǎn)變的。
【正文】 ighborhood of each pixel in the image for it. The wiener2 function applies a Wiener _lter (a type of linear filter) to an image adaptively, tailoring itself to the local image variance. The experimental results in Peak Signal to Noise Ratio (PSNR) are shown in Table 1. We find that the partition block size of 32 * 32 or 64 *64 is our best choice. Table 1 is for denoising image Lena, for di_erent noise levels and afixed partition block size of 32 *32 first column in these tables is the PSNR of the original noisy images, while other columns are the PSNR of the denoised images by using di_erent denoising methods. The PSNR is de_ned as PSNR = ?10 log10 Pi。j (B(i。 j) ? A(i。 j))2 n22552 : where B is the denoised image and A is the noisefree image. From Table 1 we can see that ComRidgeletShrink outperforms VisuShrink, the ordinary RidgeletShrink, and wiener2 for all cases. VisuShrink does not have any denoising power when the noise level is low. Under such a condition, VisuShrink produces even worse results than the original noisy images. However, ComRidgeletShrink performs very well in this case. For some case, ComRidgeletShrink gives us about dB improvement over the ordinary RidgeletShink. This indicates that by bining the dualtree plex wavelet into the ridgelet transform we obtain signi_cant improvement in image denoising. The improvement of ComRidgeletShrink over V isuShrink is even more signi_cant for all noisy levels and testing images. Figure 1 shows the noise free image, the image with noise added, the denoised ima ge with VisuShrink, the denoised image with RidgeletShrink, the denoised image with ComRidgeletShrink, and the denoised image with wiener2 for images Lena, at a partition block size of 32*32 pixels. It can be seen that ComRidgeletShrink produces visually sharper denoised images than V isuShrink, the ordinary RidgeletShrink, and wiener2 filter, in terms of higher quality recovery of edges and linear and curvilinear features. 4 Conclusions and Future Work In this paper, we study image denoising by using plex ridgelets. Our plex ridgelet transform is obtained by performing 1D dualtree plex wavelet onto the Radon transform coe_cients. The Radon transform is done by means of the projectionslice theorem. The approximate shift invariant property of the dualtree plex wavelet transform makes the plex ridgelet transform an excellent choice for image denoising. The plex ridgelet transform provides nearideal sparsity of representation for both smooth objects and objects with edges. This makes the thresholding of noisy ridgelet coe_cients a nearoptimal method of denoising for Gaussian white noise. We test our new denoising method with several standard images with Gaussian white noise added to the images. A very simple hard thresholding of the plex ridgelet coe_cients is used. Experimental results show that plex ridgelets give better denoising results than VisuShrink, wiener2, and the ordinary ridgelets under all experiments. We suggest that ComRidgeletShrink be used for practical image denoising applications. Future work will be done by considering plex ridgelets in curvelet image denoising. Also, plex ridgelets could be applied to extract invariant features for pattern recognition.
畢業(yè)設(shè)計相關(guān)推薦
鄂ICP備17016276號-1