freepeople性欧美熟妇, 色戒完整版无删减158分钟hd, 无码精品国产vα在线观看DVD, 丰满少妇伦精品无码专区在线观看,艾栗栗与纹身男宾馆3p50分钟,国产AV片在线观看,黑人与美女高潮,18岁女RAPPERDISSSUBS,国产手机在机看影片

正文內(nèi)容

畢業(yè)設(shè)計(jì)基于圖像紋理性質(zhì)的圖像修復(fù)研究(已修改)

2024-12-18 08:10 本頁面
 

【正文】 I 目 錄 摘 要 ...................................................................... Ⅲ Abstract ................................................................... Ⅴ 第一章 緒論 ............................................................. 1 前言 ............................................................ 1 相關(guān)技術(shù)研究 .................................................... 1 圖像去噪技術(shù)的研究 ......................................... 1 圖像修復(fù)技術(shù)的研究 ......................................... 2 紋理合成技術(shù)的研究 ......................................... 5 圖像修復(fù)之紋理合成技術(shù)的研究 ............................... 6 論文的主要內(nèi)容 .................................................. 7 第二章 圖像去噪方法簡介 ................................................. 9 數(shù)字圖像的矩陣表示 .............................................. 9 空間域圖像去噪 .................................................. 9 均值濾波法 ................................................ 11 中值濾波法 ................................................ 13 頻率域圖像去噪 ................................................. 14 理想低通濾波 .............................................. 14 巴特沃斯濾波器 ............................................ 15 本章小結(jié) ....................................................... 16 第三章 紋理合成的研究與分析 ............................................ 17 紋理合成的定義 ................................................. 17 紋理合成的模型 ................................................. 17 紋理合成技術(shù) ................................................... 18 紋理映射 .................................................. 18 過程紋理合成 (Procedural Texture Synthesis) ................ 21 基于樣本的紋理合成 ........................................ 22 II 本章小結(jié) ....................................................... 32 第四章 圖像修復(fù)之紋理合成技術(shù)的研究 .................................... 33 紋理修復(fù)的問題描述 ............................................. 33 圖像修復(fù)之紋理合成與圖像細(xì)紋修復(fù)的區(qū)別 .................... 33 圖像修復(fù)之紋理合成與純紋理合成的區(qū)別 ...................... 34 算法流程 ....................................................... 34 算法的完整流程 ............................................ 34 基于紋理塊的圖像修復(fù)算法流程 .............................. 35 算法實(shí)現(xiàn)方法 ................................................... 36 填充順序邊界點(diǎn)的優(yōu)先級(jí)計(jì)算 ................................ 37 尋找匹配紋理 .............................................. 38 更新像素點(diǎn)的置信度 ........................................ 39 算法的詳細(xì)描述 ................................................. 40 結(jié)果分析 ....................................................... 41 本章小結(jié) ....................................................... 41 第五章 結(jié)論與未來工作展望 .............................................. 43 研究工作總結(jié) ................................................... 43 研究工作展望 ................................................... 43 參考文獻(xiàn) ............................................................... 45 致謝 ....................................................................... 47 III 基于圖像紋理性質(zhì)的圖像修復(fù)研究 摘 要 圖像修復(fù)技術(shù)起源于文藝 復(fù)興時(shí)期,那時(shí)人們對(duì)早期中世紀(jì)十分珍貴的藝術(shù)珍品進(jìn)行修復(fù),其目的在于通過填補(bǔ)一些因?yàn)闀r(shí)間的侵蝕而造成的裂縫來使畫面恢復(fù)原貌,隨著時(shí)代的變遷,這種技術(shù)已經(jīng)不僅僅限制在修補(bǔ)古代的畫卷了,人們很自然的把這種技術(shù)過渡到了照片和電影膠片上,現(xiàn)在,人們進(jìn)行修復(fù)的目的不僅僅是為了修復(fù)一些劃痕或者污漬,而擴(kuò)展到可以在圖像上面增加或去除物體,例如電影特技的拍攝。在現(xiàn)代計(jì)算機(jī)視覺中,人們使用數(shù)字技術(shù)來完成這一工作。本文主要針對(duì)圖像細(xì)紋修復(fù)技術(shù)、紋理合成技術(shù)進(jìn)行研究,并針對(duì)較大面積的圖像修復(fù)技術(shù)進(jìn)行了理論研究和算法設(shè)計(jì)。 在圖像 細(xì)紋修復(fù)方面,本文首先對(duì)圖像細(xì)紋修復(fù)技術(shù)進(jìn)行了深入的分析和研究:圖像細(xì)紋修復(fù)主要在于根據(jù)圖像中到達(dá)破損區(qū)域邊界上的等值線來恢復(fù)圖像的色度信息,但是不能恢復(fù)圖像的紋理信息。根據(jù)較大面積修復(fù)的特點(diǎn),改進(jìn)了基于偏微分方程的圖像基本色度信息修復(fù)算法。算法只要提供待修復(fù)圖像以及需要填充圖像的掩碼圖像即可,中間不需要用戶交互,能夠自動(dòng)完成修復(fù)。 在紋理合成方面,論文首先對(duì)國內(nèi)外的各種紋理合成技術(shù)做了較為全面、系統(tǒng)的分析和闡述:然后重點(diǎn)研究了基于樣本的紋理合成技術(shù),這種技術(shù)是三大類紋理合成技術(shù)起步最晚的一個(gè),是當(dāng)今紋理合 成技術(shù)的研究熱點(diǎn)。 最后在大面積紋理修復(fù)方面,設(shè)計(jì)出了基于紋理合成和圖像細(xì)紋修復(fù)相結(jié)合的算法。在圖像修復(fù)中,僅僅使用紋理合成的方法用來修復(fù)圖像紋理信息,在匹配紋理過程中會(huì)出現(xiàn)不穩(wěn)定的匹配,針對(duì)此種現(xiàn)象,在進(jìn)行紋理修復(fù)前使用改進(jìn)后的圖像細(xì)紋修復(fù)算法對(duì)破損區(qū)域的基本顏色信息做出估計(jì),在此估計(jì)之上進(jìn)行紋理匹配。實(shí)驗(yàn)結(jié)果表明,該算法具有效果好和簡單易行的優(yōu)點(diǎn)。 關(guān)鍵詞 : 圖像修復(fù),圖像去噪,紋理合成,細(xì)紋修復(fù) IV V Study on Image Restoration Which Is Based on the Nature of the Image Texture Abstract The image restoration originated in the Renaissance when people began to restored those precious medieval artworks. The aim of this technique is to bring this damaged picture “up to date” by filling the gap in the image. With time passed, this kind of technique is not limited in repairing the medieval pictures. People transferred this technique into movie and photos. Now, the aim of restoration is not only to repair some scratches and dust spots, but also to add and delete the objects in the picture. For example, the special technique in movie is one of them. In modern puter vision, people use digital technology to plete this task. In this paper, we mainly research the image inpainting technique and texture synthesize technique and put forward an algorithm for restoring relatively big area in the image. In the image inpainting, we first make a deep research and analysis in image inpainting technique. Image inpainting mainly restored color information by pute the isophote on the edge of the damaged area. But the disadvantage of this algorithm is that it can’t restore texture information. Based on the feature of restoring relatively big area, we improve the PDE based image inpainting algorithm to repair the basic color information of the damaged image. The algorithm only needs the damaged image and the mask image. In the middle of process, it needn’t user interaction and can achieve the restoration. In the texture synthesize, we first make a plete, systematic analysis and research of the overall texture synthesizing technique all over the world. Then we mainly focus on patchbased texture synthesize. Patchbased synthesize is the newest synthesizing method in the three synthesizing methods. Patchbased is also the hottest research area in the p
點(diǎn)擊復(fù)制文檔內(nèi)容
公司管理相關(guān)推薦
文庫吧 www.dybbs8.com
公安備案圖鄂ICP備17016276號(hào)-1