影像感測器離散失真修補方法之研究

Study of the discretization effects in digital imaging

黃少宏(Shao-Hung Huang)

指導教授 ： 田仲豪

國立交通大學/電機學院/顯示科技研究所/碩士(2014年)

https://doi.org/10.6842/NCTU.2014.00691

摘要

影像感測器將光場分佈轉換為數位資料，因此在數位光學中扮演關鍵角色。我們期望良好的影像擷取品質，然而影像感測器離散化過程時常對影像品質造成各種程度的影響。使用更高性能的影像擷取設備有時可以解決問題，但是在某些情況下，這個選項因為過於昂貴或者實務上的限制而不能被實行。
本文首先討論兩種離散化過程─取樣與量化，並討論它們帶來的影響。接著，我們將討論兩種基於融合多張影像的演算法：針對取樣問題的超解析影像合成，以及針對量化問題的高動態範圍影像合成，藉此在不更動影像感測器硬體的情況下，消除或減少取樣與量化帶來的影響。

關鍵字

數位超解析 ； 高動態範圍 ； 影像合成

並列摘要

Image sensors are used to convert light intensity distribution into digital counts, which plays the important role of digital optics. However, the digital images are often degraded due to the discretization processes. Image acquisition systems with hardware upgrade are usually costly while the improvements are limited.
In this article, we investigated two discretization processes: sampling and quantization, and their influence on the image quality. In order to overcome quantitative flaws caused by discretization processes, we employed image fusion algorithms to ease or prevent the degradation problems from the discretization processes.

並列關鍵字

Superresolutopn ； HDR imaging ； Image fusion

參考文獻

[1] R. D. Fiete, Modeling the imaging chain of digital cameras. Bellingham, Wash..: SPIE, 2010.

[4] Jr. Dowski and W. Thomas Cathey, “Extended depth of field through wave-front coding,” Appl. Opt., vol. 34, no. 11, pp. 1859–1866, Apr. 1995.

[6] Patrick Vandewalle, Sabine E. Susstrunk, and Martin Vetterli, “Superresolution images reconstructed from aliased images,” 2003, vol. 5150, pp. 1398–1405.

[7] A. Papoulis, “A new algorithm in spectral analysis and band-limited extrapolation,” IEEE Trans. Circuits Syst., vol. 22, no. 9, pp. 735–742, 1975.

[9] S. Borman and R. L. Stevenson, “Super-resolution from image sequences-a review,” in 1998 Midwest Symposium on Circuits and Systems, 1998. Proceedings, 1998, pp. 374–378.