透過您的圖書館登入
IP:3.138.154.6
  • 學位論文

應用改良式立體視覺演算法於自動對焦技術之研究

A Study on the Autofocus Technology by Applying the Modified Stereovision Algorithm

指導教授 : 陳冠宇
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


由於大部份的影像監視系統常常受到攝影機解析度不足、目標距離變化及環境光源的影響,導致記錄的影像品質不佳,或遺漏關鍵畫面,甚至無法清楚辨識特徴,雖有監視設備,卻無法提供有參考價值的影像。本文的研究目的在發展一部由雙攝影機構成之立體視覺影像追蹤系統,改善上述的缺點。第一,本文提出一全新概念的自動對焦方法,乃結合改良式無焦距因素立體視覺演算法,及基於影像清晰度的開放空間域自動對焦演算法的影像伺服功能,第二,此方法可應用於移動目標物追蹤,且可呈現最佳影像畫質。,其中,結合改良式無焦距因素立體視覺演算法,更能精準的求出移動目標物,與攝影機間的距離,可減少攝影機的放大倍率,及縮小對焦點的搜尋範圍的誤差;而基於影像清晰度的對焦點演算法搭配多重閥值顏色篩選,提高所需之影像清晰度函數,再利用實測判定所制定的感測距離景深搜尋,及感測距離粗略搜尋間隔序列數,即可快速預估較佳對焦點位置,再以此位置制定一小區間,配合全域搜尋法,即得最佳對焦點位置,以記錄清晰且適度放大目標物的影像,有效增進自動對焦技術,避免對焦失誤的缺點。

關鍵字

自對對焦

並列摘要


Since most video surveillance systems are often inadequate by camera resolution, target distance change and environmental impact of the light source, resulting in poor image quality recording, or omissions key frames, and even Feature can not be clearly identified, although monitoring equipment, but can not provide a valuable reference images. This study aims to develop a constituted by a pair of stereoscopic video camera tracking system to improve the above-mentioned drawbacks. First, this paper proposes a new concept of the auto-focus method is combined with non-focal length factor improved stereo vision algorithms and image clarity of open space domain image servo autofocus algorithm functions, and the second, this method can be applied to moving target tracking, and can present the best image quality. Which, combined with improved stereo vision algorithms without focal factors, more accurate moving object is obtained, and the distance between the camera can reduce the magnification of the camera, and the narrow focus of the search for an error; while image-based definition focus algorithm with multiple threshold color filters to improve image clarity required function, re-use search depth sensing distance and sensing distance cursory search spacer sequence number, you can quickly estimate better focus position, then this position to develop an inter-cell, with the global search method, that was the best focus position to record clear and moderately enlarged target image, the effective promotion auto-focus technology, to avoid the shortcomings of focusing errors.

並列關鍵字

auto-focus

參考文獻


[3] 鄭芳炫、毛新惟,快速影像式自動對焦技術研究,Journal of Information Technology and Applications,3期1卷,頁次67-76,2008。
[9] 楊松輯,自主移動機器人具自動化對焦功能之立體視覺影像追蹤,碩士論文,中原大學機械學系,2012。
[12] 張文龍,自主移動機器人之人機互動控制的設計與實現,碩士論文,中原大學機械學系,2007。
[1] Jie He, Rongzhen Zhou, Zhiliang Hong, “Modified FastClimbing Search Auto-focus Algorithm with Adaptive StepSize Searching Technique for Digital Camera,” IEEE Trans.on Consumer Electronics,Vol. 49, pp.257-262, May 2003.
[2] G. Yang, B.J. Nelson, “Wavelet-Based Auto-Focusing andUnsupervised Segmentation of Microscopic Images,” IEEEInternational Conference on Intelligent Robots and Systems, pp. 2143-2148, 2003.

延伸閱讀