基于稀疏特性的圖像恢復(fù)和質(zhì)量評(píng)價(jià)研究
發(fā)布時(shí)間：2018-11-05 11:19
【摘要】：視覺(jué)最為人類(lèi)獲取外部信息最主要的渠道之一,對(duì)人們感知和理解外部世界起到十分重要的作用。隨著多媒體技術(shù)和傳感器技術(shù)的飛速發(fā)展,圖像對(duì)人們的生產(chǎn)、生活等產(chǎn)生越來(lái)越重要的影響。近十年以智能手機(jī)為代表的便攜智能設(shè)備變得日益普及,人們可以比歷史上任何時(shí)期更方便地記錄視覺(jué)信息。然而,由非專(zhuān)業(yè)的人士或非專(zhuān)業(yè)設(shè)備獲得的圖像不可避免地存在各種各樣的失真,從而造成人視覺(jué)感知體驗(yàn)的下降,甚至導(dǎo)致圖像語(yǔ)義信息的破壞。在絕大多數(shù)時(shí)間,人們總是傾向于得到一個(gè)清晰、鋒利、無(wú)噪聲的高質(zhì)量圖像。圖像恢復(fù)旨在濾除失真圖像中的失真部分,從而達(dá)到提升圖像質(zhì)量的目標(biāo)。在圖像恢復(fù)的過(guò)程中,一個(gè)很重要的問(wèn)題就是如何定義圖像的感知質(zhì)量。圖像質(zhì)量評(píng)價(jià)算法旨在通過(guò)計(jì)算機(jī)算法來(lái)模擬人眼視覺(jué)系統(tǒng)(Human Visual System, HVS)對(duì)圖像質(zhì)量的感知,實(shí)現(xiàn)與人類(lèi)感知一致的圖像質(zhì)量評(píng)價(jià)。圖像恢復(fù)和圖像質(zhì)量評(píng)價(jià)既相互聯(lián)系又相互區(qū)別。當(dāng)圖像遭受失真污染,如果嘗試將該失真濾除,就是圖像恢復(fù)問(wèn)題：如果嘗試評(píng)估該失真給人眼造成的質(zhì)量感知變化,就是圖像質(zhì)量評(píng)價(jià)問(wèn)題。因此,絕大多數(shù)圖像恢復(fù)算法需要以圖像質(zhì)量為風(fēng)向標(biāo),而優(yōu)秀的圖像質(zhì)量評(píng)價(jià)算法可以為圖像恢復(fù)算法提供十分有效的指導(dǎo)信息。本課題以圖像自有的稀疏特性為切入點(diǎn),針對(duì)圖像恢復(fù)和圖像質(zhì)量評(píng)價(jià)展開(kāi)深入研究,具體包括：高速運(yùn)動(dòng)車(chē)牌去模糊,視頻混合噪聲去噪和圖像模糊／鋒利程度評(píng)價(jià)。本論文的主要工作和創(chuàng)新點(diǎn)可以總結(jié)為以下幾點(diǎn)：1.論文提出了一種針對(duì)汽車(chē)高速運(yùn)動(dòng)導(dǎo)致的車(chē)牌模糊的魯棒去模糊算法。首先,根據(jù)相機(jī)的成像原理及汽車(chē)的運(yùn)動(dòng)規(guī)律,將造成車(chē)牌模糊的卷積核簡(jiǎn)化為線性卷積核。從而,卷積核的估計(jì)問(wèn)題可以簡(jiǎn)化為參數(shù)估計(jì)問(wèn)題。通過(guò)稀疏字典學(xué)習(xí),將清晰車(chē)牌圖像的先驗(yàn)信息融合在稀疏字典中,發(fā)現(xiàn)去卷積結(jié)果的稀疏表達(dá)系數(shù)與某些卷積核參數(shù)之間存在著擬凸關(guān)系,利用此性質(zhì)可以比較魯棒地估計(jì)出卷積核參數(shù),從而得到較好的車(chē)牌去卷積效果,為后面的車(chē)牌識(shí)別奠定基礎(chǔ)。2.本文提出了一種針對(duì)視頻混合噪聲去噪的非局部算法。通過(guò)分析視頻數(shù)據(jù)及噪聲數(shù)據(jù)不同的特性,利用了視頻數(shù)據(jù)在當(dāng)前幀和周?chē)鷶?shù)幀之間很強(qiáng)的自相關(guān)性。此外,視頻數(shù)據(jù)有清晰的結(jié)構(gòu)信息,其梯度分布符合一定的統(tǒng)計(jì)規(guī)律。本文從這兩個(gè)不同的特性入手,對(duì)視頻數(shù)據(jù)和噪聲數(shù)據(jù)施加不同的特性約束,利用優(yōu)化理論及方法,通過(guò)求解優(yōu)化問(wèn)題實(shí)現(xiàn)視頻數(shù)據(jù)和噪聲數(shù)據(jù)的分離,從而達(dá)到去噪的效果。3.本文提出了一種基于稀疏表達(dá)的圖像模糊／鋒利程度評(píng)價(jià)算法。圖像的結(jié)構(gòu)信息對(duì)于人的視覺(jué)質(zhì)量感知起著十分重要的作用,因此如何描述圖像結(jié)構(gòu)信息是圖像質(zhì)量評(píng)價(jià)中的一個(gè)重要問(wèn)題。通過(guò)稀疏字典學(xué)習(xí),得到的稀疏字典項(xiàng)具有清晰的結(jié)構(gòu)信息,這為使用稀疏表達(dá)進(jìn)行圖像質(zhì)量評(píng)價(jià)奠定了基礎(chǔ)。此外,通過(guò)構(gòu)建多層金字塔,克服稀疏表達(dá)無(wú)法捕獲跨尺度信息的缺點(diǎn),利用最大化池化壓縮稀疏表達(dá)系數(shù)的維度,從而實(shí)現(xiàn)圖像模糊／鋒利程度預(yù)測(cè)。本文充分利用圖像(視頻)數(shù)據(jù)的自有稀疏特性,針對(duì)圖像恢復(fù)和圖像質(zhì)量評(píng)價(jià)等多個(gè)典型問(wèn)題,設(shè)計(jì)了更加有效的圖像恢復(fù)和質(zhì)量評(píng)價(jià)算法,并深入了分析圖像的稀疏特性,大量的實(shí)驗(yàn)結(jié)果表明了本文所提算法的有效性。
[Abstract]:Vision is one of the most important sources of external information and plays a very important role in people's perception and understanding of the outside world. With the rapid development of multimedia technology and sensor technology, the image has a more and more important influence on people's production and life. portable smart devices represented by smart phones have become increasingly popular in the past decade, and visual information can be recorded more conveniently than in any period of history. However, images obtained by non-professional or non-professional devices inevitably suffer from a variety of distortions, resulting in a decrease in human visual perception experience, or even the destruction of image semantic information. In most of the time, people tend to get a clear, sharp, noiseless high-quality image. The image restoration is intended to filter out distortion parts in the distorted image, thereby achieving the goal of improving image quality. In the process of image restoration, an important problem is how to define the perceived quality of an image. The image quality evaluation algorithm is designed to simulate human visual system (HVS) perception of image quality by computer algorithm to realize image quality evaluation consistent with human perception. Image restoration and image quality evaluation are both interrelated and different from each other. When the image is subjected to distortion pollution, if the distortion is attempted to be filtered out, it is an image restoration problem: if the quality perception change caused by the distortion to the human eye is attempted, the image quality evaluation problem is solved. Therefore, most of the image restoration algorithms need to be based on the image quality, and the excellent image quality evaluation algorithm can provide very effective guidance information for the image restoration algorithm. Aiming at image restoration and image quality evaluation, this paper focuses on image restoration and image quality evaluation, including: high speed motion license plate deblurring, video mixed noise de-noising and image blur/ sharpness evaluation. The main work and innovation points of this thesis can be summarized as follows: 1. In this paper, a fuzzy algorithm is proposed to blur the license plate blur caused by high-speed motor vehicle movement. First, according to the imaging principle of the camera and the motion law of the automobile, the convolution kernel which causes the license plate blur is simplified into a linear convolution kernel. Therefore, the estimation problem of the convolution kernel can be simplified to the parameter estimation problem. Through the sparse dictionary learning, the prior information of the clear license plate image is fused in the sparse dictionary, the quasi-convex relation exists between the sparse expression coefficient of the deconvolution result and some convolution kernel parameters, the convolution kernel parameter can be estimated by using the property, so that a better license plate deconvolution effect is obtained, and a foundation is laid for the following license plate identification. This paper presents a non-local algorithm to de-noising video mixed noise. By analyzing the different characteristics of the video data and the noise data, the self-correlation between the current frame and the surrounding number frame is very strong. In addition, the video data has clear structural information, and its gradient distribution is consistent with certain statistical rules. Based on these two different characteristics, different characteristic constraints are applied to the video data and the noise data, and the optimization theory and method are used to realize the separation of video data and noise data by solving the optimization problem so as to achieve the de-noising effect. An image fuzzy/ sharpness evaluation algorithm based on sparse expression is proposed in this paper. Image structure information plays an important role in human visual quality perception, so how to describe image structure information is an important problem in image quality evaluation. With sparse dictionary learning, the obtained sparse dictionary items have clear structural information, which lays a foundation for image quality evaluation using sparse expression. In addition, by constructing multi-layer pyramid, overcoming the shortcoming that sparse expression can't capture cross-scale information, utilizing the dimension of maximizing the pool to compress sparse expression coefficient, the image blur/ sharpness prediction is realized. This paper makes full use of the self-sparse characteristics of image (video) data, designs a more effective algorithm for image restoration and quality evaluation for many typical problems, such as image restoration and image quality evaluation, and deeply analyzes the sparse characteristics of image. A large number of experiments show the validity of the proposed algorithm.
【學(xué)位授予單位】：中國(guó)科學(xué)技術(shù)大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TP391.41
[Abstract]:Vision is one of the most important sources of external information and plays a very important role in people's perception and understanding of the outside world. With the rapid development of multimedia technology and sensor technology, the image has a more and more important influence on people's production and life. portable smart devices represented by smart phones have become increasingly popular in the past decade, and visual information can be recorded more conveniently than in any period of history. However, images obtained by non-professional or non-professional devices inevitably suffer from a variety of distortions, resulting in a decrease in human visual perception experience, or even the destruction of image semantic information. In most of the time, people tend to get a clear, sharp, noiseless high-quality image. The image restoration is intended to filter out distortion parts in the distorted image, thereby achieving the goal of improving image quality. In the process of image restoration, an important problem is how to define the perceived quality of an image. The image quality evaluation algorithm is designed to simulate human visual system (HVS) perception of image quality by computer algorithm to realize image quality evaluation consistent with human perception. Image restoration and image quality evaluation are both interrelated and different from each other. When the image is subjected to distortion pollution, if the distortion is attempted to be filtered out, it is an image restoration problem: if the quality perception change caused by the distortion to the human eye is attempted, the image quality evaluation problem is solved. Therefore, most of the image restoration algorithms need to be based on the image quality, and the excellent image quality evaluation algorithm can provide very effective guidance information for the image restoration algorithm. Aiming at image restoration and image quality evaluation, this paper focuses on image restoration and image quality evaluation, including: high speed motion license plate deblurring, video mixed noise de-noising and image blur/ sharpness evaluation. The main work and innovation points of this thesis can be summarized as follows: 1. In this paper, a fuzzy algorithm is proposed to blur the license plate blur caused by high-speed motor vehicle movement. First, according to the imaging principle of the camera and the motion law of the automobile, the convolution kernel which causes the license plate blur is simplified into a linear convolution kernel. Therefore, the estimation problem of the convolution kernel can be simplified to the parameter estimation problem. Through the sparse dictionary learning, the prior information of the clear license plate image is fused in the sparse dictionary, the quasi-convex relation exists between the sparse expression coefficient of the deconvolution result and some convolution kernel parameters, the convolution kernel parameter can be estimated by using the property, so that a better license plate deconvolution effect is obtained, and a foundation is laid for the following license plate identification. This paper presents a non-local algorithm to de-noising video mixed noise. By analyzing the different characteristics of the video data and the noise data, the self-correlation between the current frame and the surrounding number frame is very strong. In addition, the video data has clear structural information, and its gradient distribution is consistent with certain statistical rules. Based on these two different characteristics, different characteristic constraints are applied to the video data and the noise data, and the optimization theory and method are used to realize the separation of video data and noise data by solving the optimization problem so as to achieve the de-noising effect. An image fuzzy/ sharpness evaluation algorithm based on sparse expression is proposed in this paper. Image structure information plays an important role in human visual quality perception, so how to describe image structure information is an important problem in image quality evaluation. With sparse dictionary learning, the obtained sparse dictionary items have clear structural information, which lays a foundation for image quality evaluation using sparse expression. In addition, by constructing multi-layer pyramid, overcoming the shortcoming that sparse expression can't capture cross-scale information, utilizing the dimension of maximizing the pool to compress sparse expression coefficient, the image blur/ sharpness prediction is realized. This paper makes full use of the self-sparse characteristics of image (video) data, designs a more effective algorithm for image restoration and quality evaluation for many typical problems, such as image restoration and image quality evaluation, and deeply analyzes the sparse characteristics of image. A large number of experiments show the validity of the proposed algorithm.
【學(xué)位授予單位】：中國(guó)科學(xué)技術(shù)大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 李文,張大鵬,劉志勇,喬香珍;跳躍與環(huán)顧最優(yōu)匹配的快速圖像恢復(fù)算法[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2002年04期
2 龍興明,周靜,馬燕;基于復(fù)數(shù)小波的圖像恢復(fù)算法研究[J];信息與控制;2004年04期
3 王創(chuàng)新;一種改進(jìn)的疊代方法圖像恢復(fù)技術(shù)[J];現(xiàn)代計(jì)算機(jī)(專(zhuān)業(yè)版);2004年07期
4 石玉英;常謙順;;權(quán)基本無(wú)振蕩格式求解圖像恢復(fù)問(wèn)題(英文)[J];計(jì)算物理;2007年02期
5 孫燮華;;關(guān)于圖像恢復(fù)算法的一點(diǎn)注記[J];中國(guó)計(jì)量學(xué)院學(xué)報(bào);2007年01期
6 孫艷麗;趙濤;于麗兵;;圖像恢復(fù)的智能處理[J];大連海事大學(xué)學(xué)報(bào);2007年S1期
7 張必銀;張?zhí)煨?劉純勝;;圖像恢復(fù)算法目標(biāo)獲取有效性的評(píng)估方法[J];紅外與激光工程;2008年01期
8 石玉英;徐靜;;關(guān)于Rudin-Osher-Fatemi圖像恢復(fù)模型特性的一個(gè)注記[J];應(yīng)用數(shù)學(xué)學(xué)報(bào);2009年03期
9 孫肖;李維國(guó);;原對(duì)偶積極集法求解改進(jìn)的有界約束圖像恢復(fù)問(wèn)題[J];中國(guó)圖象圖形學(xué)報(bào);2009年03期
10 唐曉東;劉平;;保色彩圖像恢復(fù)研究的進(jìn)展[J];中國(guó)人民公安大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年03期
相關(guān)會(huì)議論文 前6條
1 王守覺(jué);謝美芬;曹文明;;圖像恢復(fù)的一種新方法[A];2006中國(guó)控制與決策學(xué)術(shù)年會(huì)論文集[C];2006年
2 管永紅;王潤(rùn)生;施將君;;多底片圖像恢復(fù)技術(shù)研究[A];第十四屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2008年
3 柳建軍;肖庭延;王U，
本文編號(hào)：2311895
本文鏈接：http://sikaile.net/shoufeilunwen/xxkjbs/2311895.html
最近更新
教材專(zhuān)著
