Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/9309
DC FieldValueLanguage
dc.contributor賴永康zh_TW
dc.contributor.author楊俊成zh_TW
dc.contributor.authorYang, Jyun-Chengen_US
dc.contributor.other電機工程學系所zh_TW
dc.date2012en_US
dc.date.accessioned2014-06-06T06:43:03Z-
dc.date.available2014-06-06T06:43:03Z-
dc.identifierU0005-2208201202135300en_US
dc.identifier.citation[1]C. FEHN, R. D. L. BARRE, AND S. PASTOOR, “Interactive 3-DTV—concepts and key technologies“, Proceedings of the IEEE , vol.94 , no.3, pp. 524 -538 , Mar. 2006. [2]A. Redert, M. O. d. Beeck, C. Fehn, W. Jsselsteijn, M. Pollefeys, L. V. Gool, E. Ofek, I. Sexton, P. Surman, “ATTEST : Advanced three-dimensional television system technologies”, First International Symposium on 3D Data Processing Visualization and Transmission, pp. 313-319, Nov. 2002. [3] Liang Zhang, Wa James Tam, Demin Wang, “Stereoscopic image generation based on depth images“, International Conference on Image Processing , vol.5, pp. 2993 - 2996, Apr. 2004. [4]Wa James Tam, Liang Zhang,”3D-TV content generation: 2D-TO-3D conversion”, International Conference on Multimedia and Expo, pp. 1869 - 1872, Dec. 2006. [5]Liang Zhang, Senior Member, IEEE, Carlos Vazquez, Member, IEEE, and Sebastian Knorr, “3D-TV Content Creation: Automatic 2D-to-3D Video Conversion“, IEEE Trans. Broadcast. , vol.57 , no.2 , pp. 372-383, Dec. 2006. [6] J. S. Kollin, S. A. Benton, M. L. Jepsen,”Real-Time Display of 3-D Computed Holograms by Scanning the Image of an Acoustic-Optic Modulator,” SPIE Proceedings, Vol. 1212, p.174 (1990) [7]“Laser Based 3D Volumetric Display System,” US Patent No. 5,854,613 (1998) [8] C. Fehn, “3D-TV approach using depth-image-based rendering (DIBR),” in Proc. Visual., Imag. Image Process. (VIIP’03), 2003, pp. 482–487. [9]C. Fehn, “Depth-image-based rendering (DIBR), compression and transmission for a new approach on 3D-TV,” in Proc. SPIE Conf. Stereoscopic Displays and Virtual Reality Systems XI, vol. 5291, CA, U.S.A., pp. 93–104, Jan. 2004. [10] Wan-Yu Chen, Yu-Lin Chang, Shyh-Feng Lin, Li-Fu Ding, and Liang-Gee Chen , ” Efficient depth image based rendering with edge dependent depth filter and interpolation”, IEEE Trans. Consum. Electron. ,Vol. 56, No. 3, pp. 1739 - 1745 , Oct.2005 . [11] Wan-Yu Chen, Yu-Lin Chang, Hsu-Kuang Chiu, Shao-Yi Chien, and Liang-Gee Chen, “Real-time depth image based rendering hardware accelerator for advanced three dimensional television system”, IEEE International Conference on Multimedia and Expo ,pp. 2069 - 2072 , Dec. 2006. [12] Q.Wei, “Converting 2D to 3D: A Survey”, Research assignment for master program media and knowledge engineering of Delft University of technology. [13] C. Fehn, P. Kauff, M. Op de Beeck, F. Ernst, W. Ijsselsteijn, M.Pollefeys, L. Vangool, E. Ofek, and I. Sexton, “An evolutionaryand optimised approach on 3D-TV,” Proc. of IBC 2002, Int.Broadcast Convention, Amsterdam, Netherlands, Sept. 2002. [14] Kaiming He, Jian Sun , Xiaoou Tang, “Single Image Haze Removal Using Dark Channel Prior”. [15] M. T. Pourazad, Student Member, IEEE, P. Nasiopoulos, Member, IEEE,and R. K. Ward, Fellow Member, IEEE,” An H.264-based scheme for 2D to 3D video conversion”, IEEE Trans. Consum. Electron., vol. 55, No. 2, May 2009. [16] Lai-Man Po, Xuyuan Xu, Yuesheng Zhu, Shihang Zhang, Kwok-Wai Cheung and Chi-Wang Ting, "Automatic 2D-to-3D video conversion technique based on depth from motion and color segmentation”, 2010 IEEE 10th International Conference on Signal Processing ,pp. 1000 - 1003 , Dec. 2010. [17] Cheng-An Chien, Cheng-Yen Chang, Jui-Sheng Lee, Jia-Hou Chang, and Jiun-In Guo, "Low complexity 3D depth map generation for stereo applications” ,2011 IEEE International Conference on Consumer Electronics, pp. 185-186 , Jan. 2011. [18] Chih-Chung Tsai, “Algorithm and Architecture Design of Vehicle Tracking for Intelligent Cruise Control System” Graduate Institute of Electronics Engineering National Taiwan University, June 2010. [19] 陳家慶, ”以電腦視覺為基礎之行車環境資訊擷取系統”, 國立東華大學電機工程學系碩士論文, 中華民國95年6月。 [20] 鄭凌軒, ”DSP-Based之車路視覺系統之研究”, 國立中山大學電機工程學系碩士論文, 中華民國94年6月。 [21] J. MacQueen, “Some Methods for Classification and Analysis of Multivariate Observations,” In Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1, 1967, pp. 281-297. [22] 陳盈璋, “應用於2D to 3D轉換之深度知覺演算法及其硬體架構設計”, 國立中興大學電機工程學系碩士論文, 中華民國101年1月。 [23] Yu-Lin Chang, Chih-Ying Fang, Li-Fu Ding, Shao-Yi Chen, and Liang-Gee Chen, ”Depth map generation for 2D-to-3D conversion by short-term motion assisted color segmentation”, IEEE International Conference on Multimedia and Expo , pp. 1958 - 1961 , Jul. 2007. [24] Te-Wei Chiang, Tienwei Tsai, Yu-Hong Lin, Mann-Jung Hsiao, ” Fast 2D to 3D conversion based on wavelet analysis”, IEEE International Conference on Systems Man and Cybernetics , pp. 3444 - 3448 , Oct. 2010. [25] Yu-Lin Chang, Wei-Yin Chen, Jing-Ying Chang, Yi-Min Tsai, Chia-Lin Lee, and Liang-Gee Chen, ”Priority depth fusion for the 2D-to-3D conversion system”, SPIE 20th Annual Symp. on Electronics Imaging 2008. [26] P. Harman, J. Flack, S. Fox, M. Dowley, ” Rapid 2D to 3D conversion”, In Proc. SPIE Vol. 4660 , Stereoscopic Displays and Virtual Reality Systems IX, pp. 78 - 86 , 2002. [27] I. Ideses, L. P. Yaroslavsky, B. Fishbain, “Real-time 2D to 3D video conversion”, Journal of Real-Time Image Processing, vol. 2, no. 1, pp. 3-9, 2007. [28] Cheng-An Chien, Cheng-Yen Chang, Jui-Sheng Lee, Jia-Hou Chang, and Jiun-In Guo, "Low complexity 3D depth map generation for stereo applications” ,2011 IEEE International Conference on Consumer Electronics, pp. 185-186 , Jan. 2011. [29] Guo-Shiang Lin, Cheng-Ying Yeh, Wei-Chih Chen, and Wen-Nung Lie, ” A 2D to 3D conversion scheme based on depth cues analysis for MPEG videos”, ICME , pp. 1141-1145 , Jul. 2010. [30] Seonyoung Lee1, Haengseon Son2, Kyungwon Min3, ” Implementation of Lane Detection System using Optimized Hough Transform Circuit”, Circuits and Systems (APCCAS), 2010 IEEE Asia Pacific Conference on, 6-9 Dec. 2010.en_US
dc.identifier.urihttp://hdl.handle.net/11455/9309-
dc.description.abstract近年來3D影像為影像處理中最熱門的話題。由於產生3D影像庫的成本過高,為了節省大量的重新錄製成本,已有許多人著手研究2D轉3D的技術,將現有的龐大2D影像庫轉成3D影像庫。我們利用構想來與行車紀錄器作結合,將行車紀錄器紀錄的2D影像3D化,大大提升使用者日後觀賞紀錄影像的愉悅度與精彩度。此外結合影像測距的方法,來取得與前車相差的距離,藉此提醒駕駛保持適當的安全距離,減少與前車相靠太近而不及反應的意外發生。 首先,我們利用車道線與路面亮度上的明顯差異經過亮度二值化過濾出可能車道線位置,再利用霍夫轉換偵測出影像中的左右車道線。然後將左右車道線的交叉點視為畫面消失點,進而決定出水平消失線,再根據這些消失線資訊對其做場景深度圖的指派來產生背景深度值。接著,利用車底陰影找出車體位置,對其展開一較大區域,對此區域作K-mean分群;統計出車體內分部群組後,再對大區塊內掃描找出車體視為前景,且根據車子位置所在賦予深度值。最後,將前景與背景融合在一起。 定位方面:先建立背景環境收集、統計資料,畫出車子在畫面中位置與真實世界距離的對應曲線,再模擬出此對應曲線,得一函數。在程式運行中找到車子位於畫面中位置後,利用此函數推出車子的真實距離,達到測距目的。 因軟體處理時間過長,故本論文針對運算量最多的區塊作硬體化來加速運算速度。硬體架構設計方面,根據演算法需求排除不必要的運算;再將複雜的三角函數運算利用建表、查表方式達成以利硬體實作、加速運算。因為硬體架構只是整個系統一區塊,須考量整個系統運作時間,所以我們提出規格為每0.2秒30張畫面1080p解析度的消失線偵測之架構電路,運作頻率為68.04MHz。本架構將除法器切成四級管線的設計,得以配合我們演算法之處理速率,並同時降低電路之最長路徑。zh_TW
dc.description.abstractIn recent years, 3D video is a hot topic in video processing. Due to the high cost of producing a 3D video library, many people begin to study 2D to 3D conversion technology to reduce a lot of re-recording cost. They use a large existing 2D video library to turn into a 3D video library. We use the idea to combine with car camera. Convert car camera’s 2D image to 3D image. It can greatly enhance the feeling and excitement of users in viewing car camera video. In addition, it combines with image processing to obtain the distance between user’s car and the car in front to remind drivers to keep of safety distance. It can reduce the traffic accident by being too close to the vehicle in front. First, luminance of the road line is significantly different from luminance of the road surface. Then we use Binarization to filter out the possible road line points. Afterwards we use the Hough Transform to detect the image of the left and right road line, the intersection of the left and right road lines which is the vanishing point of screen, to determine the horizontal vanishing line. Then the background depth value based on the information of the vanishing line is assigned to generate the background depth map. We use the shadow under the vehicle to track the position of front car. We segment the image by K-mean clustering and identify car body by statistics which shows the most count of K-mean group. Finally, merge the foreground depth map and background depth map. iii Distance Estimation procedure: 1. Establish the background environment, and collect and statistics the real distance of front car in screen. 2. Draw the corresponding curves of a car pixels distance in the screen and the real distance in world. 3. Simulate the function of the corresponding curve and find the car position in screen by our Algorithm. 4. Use the function to estimate the car''s real distance. Due to long software processing time, the thesis attempts to speed it up by implementing hardware. To realize the hardware architecture, the algorithm needs to eliminate unnecessary operations. Complex trigonometric functions are replaced by building a look-up table to facilitate hardware accelerated computing. The hardware architecture is only one part of the entire system, we must consider the operation time of the entire system. We propose a spec for the vanishing line detection circuit with 30 frames per 0.33 seconds and 1080p resolution Operating frequency at 68.04 MHz. The divider architecture is partitioned into four stages of pipeline. This way it can conform to the processing speed of our algorithm and at the same time reduce the longest path of the circuit.en_US
dc.description.tableofcontents摘要 i Abstract ii 目錄 iv 圖目錄 vi 表目錄 ix 第一章 引言 1 一、 立體視覺的產生 1 二、 3D立體影像顯示技術 3 (一) 眼鏡式 3 (二) 裸眼式 5 三、 深度資訊 6 四、 2D轉3D影像技術 7 (一) 深度資訊圖 (Depth map) 7 (二) 2D轉3D影像處理系統 8 (三) Depth Image Based Rendering(DIBR)系統 8 五、 論文組織 10 第二章 深度感知資訊相關理論與文獻 11 一、 視覺深度感知線索 11 (一) 雙眼深度線索 12 (二) 單眼深度線索 13 二、 2D轉3D相關文獻介紹 16 (一) An H.264-based Scheme for 2D to 3D Video Conversion[15] 16 (二) AUTOMATIC 2D-TO-3D VIDEO CONVERSION TECHNIQUE BASED ON DEPTH-FROM-MOTION AND COLOR SEGMENTATION[16] 17 (三) Low Complexity 3D Depth Map Generation For Stereo Applications[17] 18 第三章 應用於行車紀錄器之2D to 3D深度圖生成與測距及其模擬結果 19 一、 前言 19 二、 車道偵測 20 三、 消失線偵測by Hough Transform 22 四、 深度背景產生 26 五、 道路延伸 27 六、 車輛陰影偵測 29 (一) 前言 29 (二) 車輛陰影特性分析 29 (三) 車輛陰影偵測演算法 31 七、 車輛位置偵測 34 (一) 車底位置偵測 34 (二) 車體左右邊界偵測 38 (三) 車頂邊界偵測 39 八、 前景產生 40 (一) K-mean 40 (二) Body Detection 43 (三) Hole Filling 44 九、 深度圖融合 45 十、 模擬結果與比較 46 (一) 模擬結果 47 (二) 結果比較 52 十一、 距離估測 55 (一) 距離估測原理 55 (二) 估測結果比較 57 (三) 誤差檢討 58 (四) 結論 58 第四章 硬體架構設計與實作 59 一、 前言 59 二、 硬體規格 60 三、 消失線硬體架構設計(Vanishing Line Architecture Design) 60 四、 各單元之硬體架構設計 61 (一) 查表(Look up table) 61 (二) F_v、F_h Memory: 62 (三) Vanishing Region Accumulator Module 62 (四) Vanishing Line Decision Module 63 五、 設計挑戰 65 六、 實作結果比較 68 (一) 數位IC設計流程簡介 68 (二) 晶片規格 69 (三) Layout圖 73 第五章 結論 75 一、 結論 75 參考文獻 76zh_TW
dc.language.isozh_TWen_US
dc.publisher電機工程學系所zh_TW
dc.relation.urihttp://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-2208201202135300en_US
dc.subject2D to 3Dzh_TW
dc.subject2D to 3Den_US
dc.subject深度知覺zh_TW
dc.subject行車紀錄器zh_TW
dc.subjectdepthen_US
dc.subjectcar cameraen_US
dc.title應用於行車紀錄器於2D to 3D轉換之深度知覺與測距演算法及消失線硬體架構設計zh_TW
dc.titleDepth Perception of 2D to 3D Conversion with Distance Estimation for Car Camera and Vanishing Line Architecture Designen_US
dc.typeThesis and Dissertationzh_TW
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.openairetypeThesis and Dissertation-
item.cerifentitytypePublications-
item.fulltextwith fulltext-
item.languageiso639-1zh_TW-
item.grantfulltextrestricted-
Appears in Collections:電機工程學系所
Files in This Item:
File SizeFormat Existing users please Login
nchu-101-7099064061-1.pdf3.53 MBAdobe PDFThis file is only available in the university internal network    Request a copy
Show simple item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.