透過您的圖書館登入
IP:13.59.141.195
  • 學位論文

虛擬取樣於影像伺服追蹤系統之應用

The application of virtual sampling in visual servo tracking system

指導教授 : 陳明新

摘要


由於影像伺服追蹤系統受限於攝影機較低的取樣速度,造成整體追蹤性能不佳。在本文中,我們提出適應性預測及內插方法。藉由此方法,得到高取樣速度下目標物運動軌跡的虛擬取樣位置,並配合伺服馬達做高速度取樣的追蹤控制,以改善整體追蹤效能。 結果顯示,使用高階的適應性線性軌跡模型做預測,配合曲線內插方法可估測出較準確的物體運動軌跡。一個低取樣速度系統,在加入預測性內插後,其追蹤效能近似高取樣速度系統。

並列摘要


The tracking performance of visual servo system is limited by the low sampling rate of camera. In this thesis, we present an adaptive prediction and interpolation method. Through this method, we could get the virtual sampling values of motion trajectory of target at high sampling rate. According to the high sampling rate trajectory, the servo motor could perform high speed tracking control. The results show that we could get more accurate motion trajectory by using high order linear model for adaptive estimation. In addition, the tracking performance of low sampling rate system in cooperated with the predictive interpolation will approach that of high sampling rate system.

參考文獻


[3] 傅培耕,「即時物體追蹤之立體視覺導引自走車」,碩士論文,中
[1] 陳寬益,「即時視覺伺服追蹤系統之設計與實現」,碩士論文,國
[8] David A.Forsyth and Jean Ponce, “Computer Vision A Modern
Approach”, Pearson Education, 2003.
[10] J. Illingworth and J. Kittler, Survey of the Hough Transform,

被引用紀錄


李宗澤(2007)。應用觀測器理論實現信號預測〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU.2007.00074

延伸閱讀