跳到主要內容

臺灣博碩士論文加值系統

(18.97.9.169) 您好!臺灣時間:2025/02/09 23:38
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:張育銘
研究生(外文):Yu-Ming Chang
論文名稱:應用於1080i至循序式8K超高解析度的基於空間域與邊緣導向內插之解交錯至影像升頻技術設計與模擬
論文名稱(外文):Spatial-Domain Edge-Directed Interpolation BasedDe-interlacing to Up-scaling Technologyfor 1080i Full HD to Progressive 8K Ultra HD
指導教授:范志鵬范志鵬引用關係
指導教授(外文):Chih-Peng Fan
口試委員:高文忠吳俊霖
口試日期:2017-07-14
學位類別:碩士
校院名稱:國立中興大學
系所名稱:電機工程學系所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2017
畢業學年度:105
語文別:中文
論文頁數:125
中文關鍵詞:解交錯影像放大影像升頻
外文關鍵詞:De-interlacingUpscalingSuper-resolution
相關次數:
  • 被引用被引用:0
  • 點閱點閱:232
  • 評分評分:
  • 下載下載:14
  • 收藏至我的研究室書目清單書目收藏:0
近期科技對於顯示器的畫質與解析度要求愈來愈高,隨著高解析度Full HD、4K與未來8K UHD顯示器的日漸普及,但現階HD節目皆為1080I格式居多,例如:境外衛星頻道HBO、FOX、Discovery訊號源,國內DVB-T廣播系統,老三台,民視與公視等皆為交錯式格式;且未來新興的壓縮格式H.265 HEVC,已不再支援交錯式格式,解交錯處理與畫面升頻因此變成了一項非常重要的技術。目前在台灣所看見的數位電視HD節目,一般來說是使用NTSC系統的1080I交錯式畫面,為了要讓交錯式畫面能顯示在循序式高解析度4K 或是8K平面顯示器,且得到更好的畫質,就得依賴解交錯演與升頻演算法的處理。本論文內容為提出一個基於低複雜度空間域之解交錯演算法,來應用於實現影像升頻與內插解交錯的計算; 提出的演算法綜觀部分可分為兩大部分: 第一部份是研究各種影像解交錯方式並提出改良,第二部份是使用提出的解交錯演算法來實現影像升頻。
本論文的解交錯演算法採用了三條掃描線偵測演算法,在解交錯時找出影像不同方向邊緣,邊緣分別為垂直與近似水平與全水平方向;而另外提出非強烈垂直與水平邊緣的其它區間,改善了判斷錯誤的部份,畫面區分為邊緣區與其它區間兩個部份,將畫面邊緣分的更明顯; 其它區間則使用LCI (Low-complexity Interpolation)方法,都是基於低複雜度內插法所發展出來的演算法; 在近水平斜邊部份,是希望能將偵測的範圍拉大,改善補錯點、邊緣模糊與雜點,而發展出大範圍偵測-低角度內插-改良型ELA(Edge-based Line Average)與中間值之內插法。所熟悉的Full HD解析度是1920x1080,所以4K解析度是3840x2160,剛好是Full HD在水平與垂直方向解析度的2倍; 以面積(總像素)來說就是4倍,而8K則是16倍,由於目前原始8K測試影像不多見,在此使用測試原始影片4K降解析度,在水平與垂直方向降4倍解析度,降成qHD格式解析度是960x540I,再由此格式解交錯演與升頻回成4K 3840x2160P格式,來達到由1080I 到 8K相似的仿真模擬效果。
本論文提出的演算法,是在不使用移動補償處理與額外暫存影像記憶體處理的情況下,盡量去節省運算量與額外硬體,以達到高品質的解交錯與影像升頻效果。
Lately, the 4K/8K UHD streaming video displayers have applied the progressive scan format to display the video contents. However, the 4K/8K UHD streaming contents are not enough and not popular for the recent broadcast networks. Practically, the popular DVB-S system still spreads the streaming videos by the 1080i format now.

In this thesis, an effective edge-directed low-complex interpolation based method is proposed for 1080i Full HD to Progressive 8K Ultra HD video post-processing. The proposed spatial-domain post-processing technology is composed of two processing stages, where the first stage is the three-scan lines based de-interlacing process for 1080i Full HD to 1080p Full HD conversion, and the second stage is the de-interlacing based upscaling process for 1080p Full HD to 8K UHD conversion.

First, the proposed pixel detector classifies the interlaced pixels into five modes, which tend to a smooth pixel, a vertical edge, a horizontal edge, a near horizontal edge, or an uncertain status. Then the corresponding de-interlacing methods are used to interpolate the de-interlaced pixels for the edge preservation. Second, the same de-interlacing scheme is applied to the upscaling process. Simulation results demonstrate that the proposed edge-directed interpolation scheme provides better subjective results than the previous spatial-domain based methods. Moreover, by paying the cost-effective complexities, the proposed method can also achieve the acceptable visually performance for the real-time de-interlacing and upscaling applications.
誌謝..........................................................i
中文摘要......................................................ii
Abstract......................................................iii
目錄..........................................................iv
表目錄........................................................vi
圖目錄........................................................vii

第一章 解交錯與影像升頻簡介..................................1
1.1 動機背景..............................................1
1.2 視訊規格..............................................1
1.3 交錯式(Interlace)影像 與 循序式(Progressive)影像......4
1.4 解交錯原理............................................7
1.5 解交錯演算法..........................................8
1.6 影像升頻、影像放大....................................9


第二章 解交錯與升頻演算法之分析與探討.........................10
2.1 相關的解交錯演算法....................................10
2.1.1 垂直雙線性平均內插法..................................10
2.1.2 場間平均內插法........................................11
2.1.3 邊緣線平均解交錯-3點模式..............................11
2.1.4 邊緣線平均解交錯-5點模式..............................12
2.1.5 三維邊緣線平均解交錯..................................13
2.1.6 垂直時間性濾波器......................................15
2.1.7 低複雜內插法-3點模式..................................15
2.1.8 增強型邊緣線平均解交錯................................17
2.1.9 移動適應性解交錯......................................20
2.1.9.1 三張場的動態偵測.....................................23
2.1.9.2 四張場的動態偵測.....................................23
2.2 實驗過的解交錯方法....................................25
2.2.1 低複雜內插法-5點模式...................................25
2.2.2 權重式低複雜度內插法...................................26
2.2.3 增強型低複雜度內插法...................................28
2.2.4 間接邊緣的低複雜度內插法...............................30
2.2.5 基於三條掃描線解交錯法.................................32
2.3 相關影像升頻放大技術種類..............................33
2.3.1 近鄰內插法.............................................33
2.3.2 雙線性內插法...........................................34
2.3.3 雙立方內插方法.........................................35
2.3.4 邊緣導向內插法.........................................36
2.4 實驗過的影像升頻方法..................................36
2.4.1 近鄰四點平均放補點.....................................36
2.4.2 利用不同奇場擴充技術以解交錯實作升頻...................39
2.4.2.1 靠左點水平解析度擴展與解交錯升頻法...................40
2.4.2.2 雙線性水平解析度擴展與解交錯升頻法...................40
2.4.2.3 立方差水平解析度擴展與解交錯升頻法...................41
2.4.2.4 新邊緣導向內插水平解析度擴展與解交錯升頻法...........42
2.5 演算法探討............................................44

第三章 提出的演算法...........................................45
3.1 提出的解交錯至影像升頻技術架構綜觀....................45
3.2 基於三條掃描線之低複雜度空間域解交錯的改良方法........46
3.3 提出的解交錯方法實現影像升頻的演算法..................61

第四章 演算法的模擬與比較結果.................................71
4.1 PSNR效能量測方法Performance measurement...............71
4.2 PSNR效能比較結果......................................73
4.3 解交錯行程結果比較........................................77
4.4 提出的解交錯方法進行影像升頻放大結果比較..................87
4.5 邊緣區間分類統計..........................................92
4.6 硬體近似模擬運算結果..............................93

第五章 總結...................................................95

參考文獻......................................................96

附錄..........................................................100
A-1...........................................................100
A-2...........................................................106
英文參考文獻:

[1] Shyh-Feng, Yu-Ling Chang, and Liang-Gee Chen, “Motion Adaptive Interpolation with Horizontal Motion Detection for Deinterlacing”, IEEE Transactions on Consumer Electronics.Volume 49, Issue 4, pp. 1256 – 1265, Nov. 2003.

[2] Yang Yuhong, Chen Yingqi, and Zhang Wenjun, “Motion Adaptive Deinterlacing Combining with Texture Detection and Its FPGA Implementation”, Proceeding of 2005 IEEE International Workshop on VLSI Design and Video Technology, pp. 316-319, May
. 2005.

[3] Yung Yuhong, Chen Yingqi, Zhung Wenjun ,”Motion adaptive deinterlacing combining with texture detection and its FPGA implementation” , Proceedings of 2005 IEEE International Workshop on VLSI Design and Video Technology, pp. 316 – 319, 28-30, May. 2005.

[4] Yanfei Shen, Dongming Zhang, Yongdong Zhang and Jintao Li, Member IEEE, “Motion Adaptive Deinterlacing of Video Data with Texture Detection“ , Proceedings of the 2004 International Symposium on Circuits and Systems, Volume 2, pp. II - 213-16 , 23-26, May. 2004.

[5] Ching-Ting Hsu, Mei-Juan Chen and Chin-Hui Huang Department of Electrical Engineering National Dong Hwa University, Taiwan ”High Performance Spatial -Temporal De-interlacing Technique Using Interfield Information” IEEE Trans. on Consumer Electronics, Volume 2, Page, Nov. 2004.

[6] Ho Young Lee, Jin Woo Park, Sang Um Choi, Tae Min Bae, and Yeong Ho Ha, “Adaptive Scan Rate Up-Conversion System Based on Human Visual Characteristics”, IEEE Trans Consumer Elec., Volume 46, Issue 4, pp. 999-1006, Nov. 2000.

[7] Hossein Mahvash Mohammadi, Pierre Langlois, and Yvon Savaria “A Five-Field Motion Compensated Deinterlacing Method Based on Vertical Motion“, IEEE Transactions on Consumer Electronics, Volume 53, Issue3, pp. 1117 – 1124 , Aug. 2007.


[8] Pei-YinCHEN ,Memberand Yao-HsienLAI , Nonmember ”A Low-Complexity Interpolation Method for Deinterlacing”, Transactions on Information and Systems archive Volume E90-D,Issue 2 table of contents, pp. 606-608, February. 2007.

[9] S. C. Tai, C. S. Yu, and F. J. Chang Department of Electronic Engineering, National Cheng Kung Universiq, Taiwan, R.O.C “A Motion and Edge Adaptive Deinterlacing Algorithm” IEEE International Conference on Multimedia and Expo.Volume 1, pp.659 - 662 , 30-30, June. 2004.

[10] P. Brox, I. Baturone, S. Sanchez-Solano” Interlaced to progressive scan conversion Using fuzzy edge-based line average algorithm” , IEEE International Workshop on Intelligent Signal Processing,1-3 Sept. pp.10 – 15, 2005.

[11] Kenju Sugiyama and Hiroya Nakamura, “A Method of Deinterlacing with Motion Compensated Interpolation”, IEEE Trans Consumer Elec , Volume 45, no.3, pp.611-616, August. 1999.

[12] Kefei Ouyang , Guobin Shen , Shipeng Li , Ming Gu “Advanced Motion Search and Adaptation Techniques for Deinterlacing” IEEE International Conference on Multimedia and Expo,6-6 , pp:374 – 377, July. 2005.

[13] KenjiSugiyama Yoshiyuki Yamada Naoya Sagara ”Improvement of Motion Compensated Inter-Field Interpolation Method for De-Interlacing” IEEE Region 10 Conference, pp.1 – 4, Nov. 2006.

[14] Yu-LinChang,Ping-HaoWu,Shyh-FengLin,andLiang-GeeChen”Four field local motion compensated de-interlacing” , IEEE International Conference on Acoustics, Speech, and Signal Processing Volume 5 , pp: V - 253-6, 17-21 May. 2004.

[15] Gwo Giun Lee, Hsin-Te Li,Ming-Jiun Wang, and He-Yuan Lin “Motion Adaptive Deinterlacing via Edge Pattern Recognition” IEEE International Symposium on Circuits and Systems, pp.2662 – 2665, 27-30 May. 2007.

[16] Hoon Yo and Jechang Jeong “Direction oriented interpolation and its application to de-interlacing”,IEEE Transactions on Consumer Electronics, Volume.48, No. 4, Nov. 2002.

[17] Min Kyu Park, Moon Gi Kang, Kichul Nam, and Sang Gun Oh “New edge dependent deinterlacing algorithm based on horizontal edge pattern” IEEE Transactions on Consumer Electronics, Volume. 49, No. 4, Nov. 2003.

[18] J.-W. Han, J.-H. Kim, S.-H. Cheon, J.-O. Kim, and S.-J. Ko, ”A novel image interpolation method using the bilateral filter,” IEEE Transactions on Consumer Electronics, vol. 56, no. 1, pp. 175-181, 2010.


中文參考文獻:

[19] 楊順博 ”An Edge-preserving based Image Interpolation Method Using Local Extrema Filtering” 中興大學論文 2014.7.

[20] 夏湘玲 ”Motion and Pattern De-interlace Algorithm” 中原大學論文 2005.7.

[21] 何恭政 ”Novel LCI-based Motion Adaptive De-interlace Technology for Video Post-processing” 中興大學論文 2008.6.

[22] 張碧倉 ” A low-complexity and edge-preservation 2-D de-interlacing algorithm with edge detector” 中興大學論文 2011.6.

參考網頁:
[23]
https://zh.wikipedia.org/wiki/DVB
[24]
https://zh.wikipedia.org/wiki/NTSC
[25]
https://zh.wikipedia.org/wiki/%E8%B6%85%E9%AB%98%E7%95%AB%E8%B3%AA%E9%9B%BB%E8%A6%96
[26]
http://doom9.cdpa.cc/index.html?/video-basics.htm
[27]
http://www.hoyo.idv.tw/data/video.htm
[28]
http://video.ee.ntu.edu.tw/~video/homework/hw1/PSNR.pdf
[29]
http://ultravideo.cs.tut.fi/#testsequences
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top