國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 = Integrati...

國立高雄大學資訊工程學系碩士班

整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 = Integration of LIDAR and vision based approaches for textured 3D scene reconstruction

紀錄類型:	書目-語言資料,印刷品 : 單行本
並列題名:	Integration of LIDAR and vision based approaches for textured 3D scene reconstruction
作者:	黃聖文,
其他團體作者:	國立高雄大學
出版地:	[高雄市]
出版者:	撰者;
出版年:	2012[民101]
面頁冊數:	73面圖，表格 : 30公分;
標題:	Levenberg–Marquardt演算法
標題:	Levenberg–Marquardt algorithm
電子資源:	http://handle.ncl.edu.tw/11296/ndltd/10975435301060038591
附註:	參考書目：面62-64
摘要註:	隨著時代的演變與科技的進步，三維運算技術之應用越來越盛行，例如：三維重建、電影、遊戲等等。三維重建中，最精準的量測方法即為雷射測距掃描。本文首先對Velodyne雷射掃描儀進行內部參數校正，校正中使用RANSAC演算法找出最適平面，再以最適平面的資料進行Levenberg–Marquardt最佳化。最佳化後將得到新的內部參數，使用最佳化的內部參數就能得到更精確的掃描資料。為了能得具紋理的三維環境模型，必須整合雷射掃描儀與相機的資料，整合雷射掃描儀與相機遇到的問題就是雷射掃描儀與相機之間的剛性轉換，本文以線性的方式解決此問題。最後使用相機投影模型將點雲資料投射在相機影像上，就可找出深度資料點所對應的顏色資訊，以獲得具色彩紋理之三維點雲模型。 With the advancement in technology, application of 3D computing is becoming popular, such as: 3D reconstruction for movies, games and so on. One of the most accurate methods of 3D reconstruction is using laser rangefinders. In our work, we propose a method to integrate range data obtained from the LIDAR and images obtained from a camera to provide a textured 3D model of the scene.We first calibrate the intrinsic parameters of the Velodyne LiDAR and use RANSAC algorithm to find the best fit plane. The data of the plane is optimized by Levenberg–Marquardt algorithm. After the optimization, the optimal intrinsic parameters which enable us to obtain more accurate 3D range data are established.Secondly, we need to solve the rigid transformation between LiDAR and camera to determine the relationship between the range data from LiDAR and the image from the camera. This is performed in a two-stage manner. A linear method is performed to obtain the initial transformation, which is then refined by Levenberg–Marquardt algorithm.We use the acquired transformation to project 3D points onto the camera image such that we can find the depth data points corresponding to the pixels in the color image. Finally, integrating the range data and the corresponding image pixels, we can obtain a textured 3D model of the scene.

整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 = Integration of LIDAR and vision based approaches for textured 3D scene reconstruction
黃, 聖文

整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 = Integration of LIDAR and vision based approaches for textured 3D scene reconstruction / 黃聖文撰 - [高雄市] : 撰者, 2012[民101]. - 73面 ; 圖，表格 ; 30公分.
參考書目：面62-64.
Levenberg–Marquardt演算法Levenberg–Marquardt algorithm

整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 = Integration of LIDAR and vision based approaches for textured 3D scene reconstruction
LDR:03537nam0a2200277 450 001 346184
005 20170214100830.0
009 346184
010 0 $b 精裝
010 0 $b 平裝
100 $a 20170214d2012 k y0chiy05 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 整合雷射掃描與電腦視覺技術且具紋理之三維環境重建 $d Integration of LIDAR and vision based approaches for textured 3D scene reconstruction $z eng $f 黃聖文撰
210 $a [高雄市] $c 撰者 $d 2012[民101]
215 0 $a 73面 $c 圖，表格 $d 30公分
300 $a 參考書目：面62-64
314 $a 指導教授：陳佳妍博士
328 $a 碩士論文--國立高雄大學資訊工程學系碩士班
330 $a 隨著時代的演變與科技的進步，三維運算技術之應用越來越盛行，例如：三維重建、電影、遊戲等等。三維重建中，最精準的量測方法即為雷射測距掃描。本文首先對Velodyne雷射掃描儀進行內部參數校正，校正中使用RANSAC演算法找出最適平面，再以最適平面的資料進行Levenberg–Marquardt最佳化。最佳化後將得到新的內部參數，使用最佳化的內部參數就能得到更精確的掃描資料。為了能得具紋理的三維環境模型，必須整合雷射掃描儀與相機的資料，整合雷射掃描儀與相機遇到的問題就是雷射掃描儀與相機之間的剛性轉換，本文以線性的方式解決此問題。最後使用相機投影模型將點雲資料投射在相機影像上，就可找出深度資料點所對應的顏色資訊，以獲得具色彩紋理之三維點雲模型。 With the advancement in technology, application of 3D computing is becoming popular, such as: 3D reconstruction for movies, games and so on. One of the most accurate methods of 3D reconstruction is using laser rangefinders. In our work, we propose a method to integrate range data obtained from the LIDAR and images obtained from a camera to provide a textured 3D model of the scene.We first calibrate the intrinsic parameters of the Velodyne LiDAR and use RANSAC algorithm to find the best fit plane. The data of the plane is optimized by Levenberg–Marquardt algorithm. After the optimization, the optimal intrinsic parameters which enable us to obtain more accurate 3D range data are established.Secondly, we need to solve the rigid transformation between LiDAR and camera to determine the relationship between the range data from LiDAR and the image from the camera. This is performed in a two-stage manner. A linear method is performed to obtain the initial transformation, which is then refined by Levenberg–Marquardt algorithm.We use the acquired transformation to project 3D points onto the camera image such that we can find the depth data points corresponding to the pixels in the color image. Finally, integrating the range data and the corresponding image pixels, we can obtain a textured 3D model of the scene.
510 1 $a Integration of LIDAR and vision based approaches for textured 3D scene reconstruction $z eng
610 0 $a Levenberg–Marquardt演算法 $a RANSAC(Random Sample Consensus)演算法 $a 內部參數校正 $a 外部參數校正 $a 三維重建
610 1 $a Levenberg–Marquardt algorithm $a RANSAC(Random Sample Consensus) algorithm $a intrinsic calibration $a extrinsic calibration $a 3D reconstruction
681 $a 008M/0019 $b 464103 4410 $v 2007年版
700 1 $a 黃 $b 聖文 $4 撰 $3 85212
712 0 2 $a 國立高雄大學 $b 資訊工程學系碩士班 $3 353878
801 0 $a tw $b NUK $c 20121107 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/10975435301060038591