繞射光學元件應用於透明顯示器的影像品質改善

Abstract

顯示器與我們的生活息息相關，路上人人一台的手機、遠端上課所需的電腦或平板、或是客廳中的電視機，都是屬於顯示器面板產業的下游終端產品，技術持續發展，人們對顯示器的要求亦逐漸提高，除了扭曲向列型液晶、VA 垂直排列技術、IPS平面轉換技術等先進的面板技術，以及Full HD、4K UHD、8K4K等更高畫質的顯示技術外，將顯示器變透明且輕薄等也是另一項趨勢。目前透明顯示器之應用主要在商品櫥窗展示、博物館展示，因其技術特性得以使實體搭配顯示器所提供的相關產品資訊，讓資訊傳播能以更直觀的方式呈現。 目前市面上常見的透明顯示器為OLED透明顯示器，其設計原理為將畫素集中排列，沒有畫素的部分使用透明材料，透明度一般在40%以上，不透光的畫素為週期性排列，因此光通過透明顯示器時會產生繞射，亮度變暗，導致後景影像品質下降。本論文的目標是以演算法設計繞射光學元件 (diffractive optical element, DOE) ，使已產生繞射的影像通過此繞射光學元件，再以MSE、PSNR、SSIM、Correlation Coefficient、等影像評估方式去判斷影像通過此繞射光學元件是否提高了成像結果品質，並期許本研究能對開發新型透明顯示器盡一份心力。

The display is closely related to our lives. Everyone on the road with a mobile phone, a computer or tablet for remote classes, or a TV in the living room are downstream terminal products belonging to the display panel industry. Technology continues to develop. The requirements of monitors are gradually increasing. In addition to advanced panel technologies such as twisted nematic liquid crystal, VA vertical alignment technology, IPS plane conversion technology, and higher-quality display technologies such as Full HD, 4K UHD, 8K4K, etc. Transparency and thinness are another trend. At present, the application of transparent displays is mainly used in the display of merchandise windows and museums. Because of its technical characteristics, it can make the physical products match the related product information provided by the display, so that the information can be presented in a more intuitive way. At present, the common transparent displays on the market are OLED transparent displays. The design principle is to arrange the pixels centrally. The parts without pixels use transparent materials. The transparency is generally more than 40%. The opaque pixels are arranged periodically, so when light passes through a transparent display, it will be diffracted, and the brightness will become dark, resulting in a deterioration of the quality of the background image. The goal of this thesis is to design Diffractive Optical Element (DOE) algorithmically so that the diffracted image passes through the diffractive optical element, and then use MSE, PSNR, SSIM, Correlation Coefficient, and other image evaluation methods. To judge whether the image improves the quality of the imaging result through this diffractive optical element, and hope that this research can contribute to the development of a new transparent display.