用主題區域偵測完成快速自動聚焦

Fast Autofocus Using Subject Region Detection

謝銘和(Ming-He Hsieh)

指導教授 ： 傅楸善

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2006年)

https://doi.org/10.6342/NTU.2006.10134

摘要

在本論文中，我們提出了一個高速且高準確性的數位相機自動對焦的方法。
在數位相機與數位攝影機中，對焦系統的好壞會對整個影像產生很大的影響。而在對焦系統中如何計算聚焦值是在對焦系統中非常重要的一環，在傳統及現今的方法中有很多的計算方法，而且在產業中用的方法都不盡相同，在一些新的論文中有提到一些速度更快或更穩定的方法，我在此論文中聚焦值的計算是使用絕對重心要素值 (ACM: Absolute Central Moment)與高通絕對重心要素值 (HPACM: High-Pass Absolute Central Moment) ，而聚焦值的搜尋是參考張家豪的適應性搜尋步數演算法，另外使用選擇性對焦視窗來找尋最高的對焦點。
我們在此論文提供了高準確度的人物區域偵測分數系統以及利用影像特性找尋主題物體偵測的分數系統來得到主要主題格與次要主題格，進而精確及有效的來找尋最佳聚焦位置。

關鍵字

自動聚焦 ； 對焦 ； 主題偵測

並列摘要

In this thesis, we propose a high speed and high accuracy method to implement digital still camera autofocus system.
The focusing system is an important factor to determine the imaging quality of a digital still camera and video camcorder.
Moreover the image focus value evaluation algorithm is a key aspect in implementing the focusing system. Now, many kinds of auto focusing methods on the different types of digital still cameras, and various algorithms have been presented, but there are some deficiencies for these algorithms.
We provide a faster ACM (Absolute Central Moment) and HPACM (High-Pass Absolute Central Moment) for efficient focus value evaluation, adaptive auto focus search algorithm, and a new focusing windows selection method are proposed.
Our proposed high accuracy region detection with human score system, and image property score system to find the human subject or object subject region. Our method can find the primary and secondary subject region to improve the accuracy and efficiency of the traditional autofocus system.

並列關鍵字

Autofocus ； Subject Region Detection ； Fast

參考文獻

[2] C. Cavalcanti and H. Gomes, “People Detection in Still Images Based on a Skin Filter and Body Part Evidence,” Proceedings of International Symposium on Computer Graphics, Image Processing and Vision, SIBGRAPI, Natal, RN, Brazil, pp. 13039-13040, 2005.

[5] K. S. Choi, J. S. Lee, and S. J. Ko, “New Autofocus Technique Using the Frequency Selective Weighted Median Filter for Video Cameras,”IEEE Transactions on Consumer Electronics, Vol. 45, No. 3, pp. 820-827, 1999.

[8] J. He, R. Zhou, and Z. Hong, “Modified Fast Climbing Search Auto-focus Algorithm with Adaptive Step Size Searching Technique for Digital Camera,” IEEE Transactions on Consumer Electronics, Vol. 49, No. 2, pp. 257-262, 2003.

[11] F. Li and H. Jin, “A Fast Auto Focusing Method for Digital Still Camera,” Proceedings of International Conference on Machine Learning and Cybernetics, Guangzhou, Vol. 8, pp. 5001- 5005, 2005.

[13] K. Ooi, K. lzumi, M. Nozaki, and I. Takeda, “An Advanced Autofocus System for Video Camera Using Quasi Condition Reasoning,” IEEE Transactions on Consumer Electronics, Vol. 36, No.3, pp. 526-530, 1990.