【正文】 本科畢業(yè)設計 (論文 ) 題目： 基于數(shù)字圖象處理的自動對焦 技術研究 院 （系）： 電子信息工程 專 業(yè)： 電子信息工程 班 級： 學 生： 學 號： 指導教師： 2021 年 6 月I 摘要 自動對焦技術是機器人視覺、數(shù)字成像系統(tǒng)和各種精密光學儀器中的關鍵技術。隨著科學技術的飛速發(fā)展，數(shù)字成像系統(tǒng)中的精確自動對焦問題越來 越受到人們的普遍關注，其應用也顯得越來越重要。這對進一步提高測量精度、測量速度及測量自動化程度等方面具有重要的現(xiàn)實意義。 研究了影響自動對焦精度及對焦重復性的因素。針對對焦過程中存在的噪聲干擾，采用了濾波算法進行圖像預處理，實驗表明中值濾波是一種較好的方法；用對焦窗口大小、位置、對焦區(qū)域選擇等窗口規(guī)劃方法來克服對焦過程中背景因素的引入對主體目標的影響。分析表明較好的窗口規(guī)劃不僅可以一定程度解決上述問題，還可提高對焦函數(shù)性能，有效減小計算量。 在較全面地分析了基于數(shù)字圖像處理的對焦評價函數(shù)基礎上，針對基于邊緣檢測算法的對焦評價函數(shù)做了驗證性實驗，結果表明對于不同的評價目標，對焦函數(shù)的性能的精度和速度也不盡相同。通過實驗驗證，基于邊緣檢測的 laplaci an 算子，性能更好。 清晰度評價函數(shù)極點搜索算法則是實現(xiàn)對焦點的搜索和定位，通過比較評價函數(shù)值決定鏡頭移動方向，反饋控制直至成像質量最佳。在分析比較了幾種對焦技術中常用的搜索算法的優(yōu)缺點的基礎上，確定改進的爬山算法搜索在速度不減的情況下，精度更高。通過仿真驗證了該算法的有效性、可行性，可較好的實現(xiàn)自動對焦。 關鍵詞：自動對焦；對焦窗口；清晰度評價函數(shù)；極點搜索 算法 II The Research of Autofocusing System Based on Image Processing Abstract Autofocusing technology is a key technology in the robot vision and imaging System and various optical precision instruments. With the rapid development of science and technology， the autofocusing problem has been much concerned by people， and its application has been more and more can improve the accuracy and speed of measurement， automatic level . Study on the affecting factors of automatic focusing accuracy and repeatability. Considered the noise in the process of the focus， the paper used the filtering algorithms for image show median filtering is a better way； in order to over e the interference between the main image of the target and the background image， the paper used the layouts of focusing window size，location， and focusing regional selection etc. Analysis shows that better window planning not only can partly solve the problem， but also improve the focusing performance function， reduce the amount of calculation. In a more prehensive analysis of the focusing evaluation functions based on the digital image proeessing， the experiment using the functions based on the edge detection algorithms was carried out. The results show that for different evaluation target， performance focusing function of the precision and speed is not the same. Through the experiment， the Laplacian based on edge detection operators better performance. Peak searching algorithm in Sharpness evaluation function is to search the focus and location by paring the evaluation function values ， and determine the lens moving direction， feedback control and the image quality is best. After paring and analyzing the advantages and disadvantages of several mon focus search algorithm . Determine the improved hillclimbing algorithm search with higher accuracy. Key words： automatic focusing。 the focusing window。 sharpness evaluation function。 peak searching algorithm 目錄 III 目錄 中文摘要 ......................................................................................................................I 英文摘要 .................................................................................................................... II 1 緒論 .......................................................................................................................... 1 研究背景及意義 ............................................................................................... 1 自動對焦研究現(xiàn)狀 ........................................................................................... 2 本文主要研究工作 ........................................................................................... 3 2 自動對焦系統(tǒng)設計 ............................................................................................. 4 自動對焦的基本原理和方法 ........................................................................... 4 傳統(tǒng)對焦方法 ........................................................................................ 4 基于數(shù)字圖像處理的自動對焦方法 ..................................................... 5 自動對焦系統(tǒng)結構 .......................................................................................... 6 本章小結 ........................................................................................................... 7 3 影響自動對焦及對焦函數(shù)性能的因素分析 ............................................... 8 噪聲影響分析 ................................................................................................... 8 光線影響分析 ................................................................................................. 10 對焦窗口的選擇分析 ..................................................................................... 12 對焦窗口的選擇方法 ........................................................................... 12 對焦窗口的選擇策略 .......................................................................... 14 本章小結 ......................................................................................................... 15 4 自動對焦評價方法 ........................................................................................... 16 清晰度 評價函數(shù)的評價標準 ......................................................................... 16 目錄 IV 清晰度評價函數(shù)的分類 ................................................................................. 16 清晰度評價函數(shù)的改進 ................................................................................. 18 實驗和分析 ..................................................................................................... 20 本章小結 ......................................................................................................... 23 5 極點搜索算法的探討 ....................................................................................... 24 Fibonacci 搜索算法 ...................................................................................... 24 曲線擬合算法 ................................................................................................ 25 “盲人”爬山算法 ...................................................