本文選題：Contourlet變換　切入點(diǎn)：圖像水印　出處：《吉林建筑大學(xué)》2017年碩士論文

【摘要】：隨著中國經(jīng)濟(jì)高速增長,以物聯(lián)網(wǎng)和云計(jì)算為代表的信息化技術(shù)掀起了計(jì)算機(jī)、通信、信息內(nèi)容的監(jiān)測與控制的4C革命,網(wǎng)絡(luò)功能開始為社會(huì)各行業(yè)和社會(huì)生活提供全面應(yīng)用。數(shù)字電視、遙感圖像傳輸、社交媒體、醫(yī)用顯微圖像處理、機(jī)器視覺、電子商務(wù)等應(yīng)用領(lǐng)域廣泛擴(kuò)展,使得圖像數(shù)據(jù)越來越龐大,圖像的存儲(chǔ)和傳輸成為了日益增長的難題,因此,一種新型的圖像編碼技術(shù)成為了當(dāng)下最受關(guān)注的焦點(diǎn)。本文以Contourlet變換和主成分分析為基礎(chǔ),提出了一種對幾何變換和圖像魯棒壓縮的一種水印解碼器。該方案利用噪聲識別功能,自適應(yīng)調(diào)整水印嵌入強(qiáng)度,保持圖像的感知質(zhì)量,在各種可能的失真下可以高精度地檢測水印。首先,針對Contourlet變換的原理和性質(zhì),本文對Contourlet為基礎(chǔ)的圖像水印、圖像融合、圖像壓縮領(lǐng)域的應(yīng)用和發(fā)展前景進(jìn)行了深入探索和介紹,敘述了Contourlet變換的基本原理,為本文研究打下了詳實(shí)的理論基礎(chǔ)。其次,本文利用NIG作為圖像的輪廓系數(shù)來提出基于NIG分布的最大似然法的水印提取統(tǒng)計(jì)方法。使用獲得的閉式表達(dá)式分別在不存在和存在失真的環(huán)境中提取水印位置,以驗(yàn)證所提出解碼器的魯棒性。最后,結(jié)合冗余離散小波變換和小波原子變換,對于壓縮圖像算法提出了一種基于一幅圖像中的質(zhì)地和紋理具有不同的形態(tài)特征的方法。通過Contourlet變換和壓縮感知的組合重建壓縮圖像的質(zhì)地,并通過單層離散小波變換和壓縮感知的組合重建壓縮圖像的紋理。實(shí)驗(yàn)結(jié)果表明,該方法在保證原始圖像主要結(jié)構(gòu)的基礎(chǔ)上,確保了低采樣率下細(xì)節(jié)的質(zhì)量,確保了圖像質(zhì)地和紋理清晰。此外,它還具有較高的壓縮率。
[Abstract]:With China's rapid economic growth, the information technology represented by the Internet of things and cloud computing has set off a 4C revolution in the monitoring and control of computers, communications, and information content. The network function began to provide a comprehensive application for various industries and social life. Digital TV, remote sensing image transmission, social media, medical microscopic image processing, machine vision, electronic commerce and other applications have been widely expanded. As the image data becomes larger and larger, the storage and transmission of image becomes a growing problem. Therefore, a new image coding technology has become the focus of attention. This paper is based on Contourlet transform and principal component analysis. In this paper, a watermarking decoder for geometric transformation and image robust compression is proposed, which adaptively adjusts the embedding strength of the watermark and maintains the perceptual quality of the image by using the noise recognition function. The watermark can be detected with high accuracy under various possible distortions. Firstly, according to the principle and properties of Contourlet transform, this paper focuses on the image watermarking based on Contourlet and image fusion. The application and development prospect of image compression are deeply explored and introduced, and the basic principle of Contourlet transform is described, which lays a solid theoretical foundation for the research of this paper. In this paper, NIG is used as the contour coefficient of the image to propose a statistical method of watermark extraction based on the maximum likelihood method of NIG distribution. The obtained closed expression is used to extract the watermark position in the environment where there is no existence or distortion, respectively. Finally, the redundant discrete wavelet transform and wavelet atomic transform are combined to verify the robustness of the proposed decoder. For the compressed image algorithm, a method based on the texture and texture in a single image with different morphological features is proposed. The texture of the compressed image is reconstructed by the combination of Contourlet transform and compression perception. The texture of compressed image is reconstructed by the combination of single-layer discrete wavelet transform and compression perception. The experimental results show that the method ensures the quality of details at low sampling rate on the basis of guaranteeing the main structure of the original image. Ensure the texture and texture of the image clear. In addition, it also has a high compression ratio.
【學(xué)位授予單位】：吉林建筑大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 肖偉;汪榮峰;;基于非下采樣contourlet變換與脈沖耦合神經(jīng)網(wǎng)絡(luò)的圖像融合算法[J];計(jì)算機(jī)應(yīng)用;2008年S2期

2 吳艷;肖平;王昌明;李明;;基于Contourlet域持續(xù)性和聚集性的合成孔徑雷達(dá)圖像融合分割算法[J];光學(xué)學(xué)報(bào);2010年07期

3 楊昤;郭寶龍;倪偉;;基于層結(jié)構(gòu)的Contourlet多閾值圖像去噪算法[J];計(jì)算機(jī)工程;2006年20期

4 勵(lì)金祥;金煒;楊任爾;周亞訓(xùn);;合成孔徑雷達(dá)圖像的Contourlet域空間選擇濾波降斑算法[J];光學(xué)技術(shù);2009年02期

5 梁棟,沈敏,高清維,鮑文霞,屈磊;一種基于Contourlet遞歸Cycle Spinning的圖像去噪方法[J];電子學(xué)報(bào);2005年11期

6 孟慶文;周焰;徐穎;;Contourlet域的多極化SAR圖像自適應(yīng)融合方法[J];空軍預(yù)警學(xué)院學(xué)報(bào);2014年04期

7 沙宇恒,叢琳,孫強(qiáng),焦李成;基于Contourlet域HMT模型的多尺度圖像分割[J];紅外與毫米波學(xué)報(bào);2005年06期

8 楊慶;韋志輝;黃陳蓉;;基于Contourlet域的SAR圖像去噪算法[J];計(jì)算機(jī)工程;2010年21期

9 楊曉慧;趙艷敏;;基于融合策略的非下采樣Contourlet域SAR圖像去噪與仿真分析[J];系統(tǒng)仿真學(xué)報(bào);2010年01期

10 李康;陳雪軍;趙乾宏;沈宇;程立群;朱文婷;;Contourlet域自適應(yīng)萎縮閾值去噪方法[J];信息技術(shù);2012年07期

相關(guān)會(huì)議論文 前1條

1 劉浩淼;賈永紅;姚芳;;基于contourlet變換的圖像融合方法研究[A];第十七屆中國遙感大會(huì)摘要集[C];2010年

相關(guān)碩士學(xué)位論文 前1條

1 李治;基于Contourlet算法的新型圖像編碼技術(shù)研究與驗(yàn)證[D];吉林建筑大學(xué);2017年

，

本文編號：1657082