臺灣博碩士論文加值系統

本論文之研究目標之一是以低成本單眼影像、快速量測距離的方式建構整個系統為考量，影像距離量測系統使用影像輸入設備(CCD)擷取光源訊號(Super LED)，提供使用者進行對目標物的距離計算，目標之二則是模擬人體頭部動作利用雙眼視覺追蹤目標物。
單眼視覺研究之目的在設計並實現具有環境視覺深度的影像量測，實現速度快而有效的影像距離量測系統為目標。傳統的距離量測,需使用者利用捲尺來量得實際距離，一般來說，單一影像鏡頭(CCD)不容易得知目標物與影像輸入裝置之距離。本系統之硬體架構平台是使用光源(Super LED)、網路攝影機(Web camera)和凸透鏡所組成。依照光軸互相平行法架設，透過USB傳輸介面將影像輸入到系統主機上，利用光源與目標物成一直線，進而將利用光源變化的大小及影像處理之資料，導入距離量測法則便可以計算出距離。
本研究另一方式則是利用雙眼視覺來模擬人體頭部的動作，利用雙眼的焦點對正再利用單眼影像距離的演算法來套用計算出系統平台與目標物的實際距離，利用伺服馬達補正影像的誤差值，將目標物置於雙眼視覺影像的中間，而目標物如果只能利用單眼影像視覺系統觀察亦可利用單眼影像判斷距離法則來做判斷的動作。
實驗之結果可知，利用單眼系統之架構及方式，藉由視覺演算法將影像處理之資料計算後，能夠即時計算出目標物與影像鏡頭之間的距離。而雙眼影像視覺系統部分可以在一般環境中追蹤目標物移動不受光害影響，單眼影像視覺系統距離判斷及雙眼影像視覺系統距離判斷比較兩者的量測距離判斷誤差。

One of the main goals in this thesis is to construct a whole system by considering using a low cost monocular image and fast distance measurement method. The image distance system used by image input devices called CCD to acquire the light source signal (Super LED) provides users to calculate the distance of the objects. The other goal is to simulate the motion of the human being’s head by using the binocular vision to trace objects.
The purpose of this thesis is to design and explore the image measurement involved in the surrounding vision depth by using the fast and effective image distance measurement system. Traditionally, the users need to apply rulers to measure the exact distance. Generally speaking, it is not easy to apply the mono-image lens (CCD) to obtain the distance between the object and image input device. The hardware structure platform of this system consists of light source Super LED, Web Cameras and Convex Lens. It is set up according to the optical axis parallel method to transmit the image to the main engine of the system through USB transmission interface and then derive the distance measurement method through the light source variation and the information about the image treatment by applying the property of the light source and the object to be set in a line to calculate the distance we need.
The other way of this thesis is using the binocular vision effect to modify the motion of the head of man. Basically the exact distance between the system platform and the object can be solved by aligning the binocular focus and applying the algorithm which is derived to find the distance of the monocular image. This method shows that the servomotor can be applied to correct the image error while the object is set in the center of the image. If the object can be observed only by one single location, the monocular method will be also used to estimate its action.
From the experimental results we know that the distance between the object and the image lens can be instantaneously determined by adopting the structure of this monocular system and treated image data by the vision algorithm. However, for the binocular measurement, we can trace the object under general environment and make a comparison between the monocular measurement error and the binocular one.

誌謝 I
摘要 II
Abstract III
序論 IV
第一章 系統設計 1
第一節 單眼影像視覺系統設計 1
一、整體介紹 2
二、影像輸入 2
第二節 雙眼影像視覺系統設計 5
一、整體介紹 6
二、影像輸入 7
第二章 影像系統架構 9
第一節 單眼影像視覺系統之架構 9
一、硬體規格 9
第二節 雙眼影像視覺系統之架構 11
一、雙眼影像視覺系統硬體架構 11
二、8134伺服控制卡(4-Axis Motion Control Card) 16
三、硬體規格 17
第三章 影像處理 21
第一節 影像判斷法則介紹 23
第二節 單眼視覺系統影像處理 24
一、灰階化(Gray Scale) 24
二、影像二值化(Image Threshold) 25
三、GFLSDK程式開發庫(GFLSDK Library) 26
四、單眼影像視覺系統複合式判斷法則 26
五、單眼影像視覺系統距離判斷探討 30
第三節 雙眼視覺系統影像處理 34
一、影像前處理 34
二、雙眼影像視覺系統判斷探討 36
三、雙眼影像複合式判斷法則 38
四、影像合併 41
五、雙眼影像視覺深度感知計算 42
第四章 影像距離判斷法則 49
第一節 單眼影像視覺距離判斷法則 49
一、光源與凸透鏡最佳距離 49
第二節 雙眼影像視覺定位判斷法則 52
一、雙眼影像視覺系統有效距離 52
二、目標物形狀判斷 54
第五章 架構分析與實驗 56
第一節 單眼影像視覺實驗與分析 56
一、透鏡影響分析 56
二、光源與透鏡間距分析 58
三、現實環境影響分析 60
四、二維平面視覺實驗 62
五、不同影像辨識法比較 63
第二節 雙眼影像視覺定位實驗與分析 65
一、現實環境影響分析 65
二、三維平面視覺實驗 67
第六章 影像距離量測系統與現實環境之整合應用 82
第一節 單眼影像視覺系統介紹 82
一、單眼視覺系統硬體架構 82
二、濾光片對視覺影響分析 83
三、現實環境中濾光片成像影響 85
第二節 雙眼影像視覺系統介紹 86
一、雙眼影像視覺系統硬體架構 86
二、雙眼視覺系統影像定位 87
第七章 結論與成果 92
參考文獻 94

[1] Starck , L .Jean, Fionn Murtagh, J. Candes, Emmanuel, and L. Donoho, David，2003.”Gray and Color Image Contrast Enhancement by the Curve let Transform”，IEEE TRANSACTIONS ON IMAGE PROCESSING.,12(6).
[2] Olson ,Clark F. and P. Huttenlocher, Daniel，1997.”Automatic Target Recognition by Matching Oriented Edge Pixels”，IEEE TRANSACTIONS ON IMAGE PROCESSING., 6(1).
[3] Peter ,V. Henstock and M. Chelberg, David，1966.”Automatic Gradient Threshold Determination for Edge Detection”，IEEE TRANSACTIONS ON IMAGE PROCESSING.,5(5).
[4]M. John, G. Ovidiu, Paul F. Whelan，”Robust 3-D Landmark Tracking using Trinocular Vision”，Vision System Laboratory, School of Electronic Engineering Dublin City University.
[5]K. Joss，”Robot navigation by active stereo vision”，Active Vision Laboratory.
[6] Massimo, Berozzi, B. and Alberto，1988.”GOLD: A Parallel Real-Time Stereo Vision System for Generic Obstacle and Lane Detection”， IEEE TRANSACTIONS ON IMAGE PROCESSING. ,7(1).
[7] Yau, W. Y. and H.,Wang,1999. “Fast Relative Depth Computation for an Active Stereo Vision System,” Real-Time Imaging, 5(3): 189-202.
[8] A. Ohya, K. Akio, and K. Avinash,1996 “Vision-Based Navigation by a Mobile Robot with Obstacle Avoidance Using Single-Camera Vision and Ultrasonic Sensing ”.Vision Res, 36(23).
[9] S.T. Barnard, and W.B. Thompson,,1980. “Disparity Analysis of Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2:(4),330-340.
[10] Tsai, R. Y.,1987 “A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-Shelf TV Cameras and Lenses,” IEEE J. of Robotics and Automation, RA-3(4): 323-344.
[11] Sun, G. Z. et al.,1988 “Learning Stereopsis with Neural Networks,” Proceedings of the International Joint Conference on Neural Networks, San Diego, 4:345-355.
[12] Mohan, R. and R. Nevatia,1989 “Using Perceptual Organization to Extract 3D Structures,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 11(11): 1121-1139.
[13] Marichal,G.. N., L. Acosta ,L. Moreno, J. A. Méndez, J. J. Rodrigo, and M. Sigut,2001. “Obstacle Avoidance for a Mobile Robot: A Neuro-Fuzzy Approach,” Fuzzy Sets and Systems, 124(2): 171-179.
[14] Han, S.-H., W. H. Seo, K. S. Yoon, and M.-H. Lee,1999 “Real-Time Control of an Industrial Robot Using Image-Based Visual Servoing,” Proceedings of the 1999 IEEE/RSJ International Conference on Intelligent Robots and Systems,(3):1762-1796.
[15] Winters, N. and J. Santos-Victor,2002 “Information Sampling for Vision-Based Robot Navigation, ” Robotics and Autonomous Systems, 41(2-3):145-159.
[16]Chen, K. Y., Chang, Y. P., and Fu, B.,2003 “Design and Implementation of a Real-Time Image Tracking System,” Chung Yuan Journal, 31(1):21-29.
[17] Kuo ,H.C. and L.J. Wu,2002 “An image tracking system for welded seams using fuzzy logic,” Journal of Materials Processing Technology, 120(1-3):69-185.
[18] Shigang Yuea, F. Claire Rinda, S. Keilb. Matthais, Cuadrib. Jorge and Richard Staffor.2005“ A bio-inspired visual collision detection mechanism for cars” Received 28 October 2004; received in revised form 31 (6).
[19] M. Bj¨orkman, D. Kragic, 2004“Combination of foveal and peripheral vision for object recognition and pose estimation,”Proceedings of the IEEE International Conference on Robotics and Automation, RA-4( 5): 5135–5140.
[20] Kim ,S., Kim ,I. and I. Kweon ,2003, “Robust model-based 3d object recognition by combining feature matching with tracking”, Proceedings of the IEEE International Conference on Robotics and Automation, ICRA’, 2:2123–2128.
[21] Seara ,J.F.and G. Schmidt,2004 “Intelligent gaze control for vision-guided humanoid walking: methodological aspects” Robotics and Autonomous Systems 48(4): 231–248.
[22] K. Kidono, J. Miura, Y. Shirai,2002“Autonomous visual navigation of a mobile robot using a human-guided experience” Proceedings of the Elsevier International Conference on Robotics and Autonomous Systems 40: 121–130.
[23] J. Lobo, C. Queiroz, J. Dias,2003“World feature detection and mapping using stereovision and inertial sensors” Proceedings of the Elsevier International Conference on Robotics and Autonomous Systems 44 :69–81.
[24]徐國銘，2003“低成本影像追蹤系統之研究開發與應用”，大同大學碩士論文。
[25]戴呈霖，2002,“雙照相機座標量測系統”，國立成功大學碩士論文。
[26] 傅培耕，2002“即時物體追蹤之立體視覺導引自走車”，中原大學碩士論文。