微型鏡頭陣列之高動態範圍影像技術

Abstract

隨著時代進步，科技產品朝向輕薄短小發展，鏡頭所要求的規格隨之越來越小，而微小鏡頭的機構在實際應用上面臨許多挑戰，如微小鏡頭進光量較一般鏡頭少而導致雜訊增加、降低影像品質。另外，目前常見的影像感測器無法完整將真實場景中的動態範圍保存下來。利用多張不同曝光影像合成單一高動態範圍影像的技術能夠克服上述問題。 高動態範圍影像處理相關技術已相當成熟，但如何控制曝光以選擇適當來源影像較少被討論，因此本論文提出了一以影像亮度直方圖為基礎的自動多重曝光選擇技術，能針對不同場景，自動調整每個鏡頭各自的曝光時間，讓影像中每個像素值都能有兩個以上的有效參考值，搭配所提出的權重函式，有效抑制微型鏡頭陣列所造成的隨機分布雜訊，進一步生成高品質的高動態範圍影像。 取得高動態範圍影像後，須將影像壓縮至低動態範圍以供一般顯示器播放，即色調對應的技術。本論文提出了利用全域壓縮和影像混和的技術，有效將高動態範圍影像壓縮至低動態範圍，供顯示器顯示並保留場景中細節。本論文利用四顆微型鏡頭陣列實現所提演算法，先將所取得的來源影像對齊至同一平面之後，經所提出之微型鏡頭陣列之高動態範圍影像演算法，針對各種場景都能夠保留暗處與亮處之細節，提高影像的動態範圍。整體系統實作在Adlink MXC-6300 (Industrial Computer)平台上，可以達到每秒約10張畫面。

Camera specification becomes smaller and smaller accompanied with great strides in technology and thinner product demands, which leads to some challenges and problems. One of those problems is micro lens captures less light than normal lens does, which makes low quality noise-image. Moreover, current image sensor cannot preserve whole dynamic range in real world. HDR image with multi-exposure image overcomes the problems mentioned above. Choosing good exposure time is a seldom-discussed but important issue in HDR imaging technology. This thesis proposes a Histogram Based Exposure Time Selection (HBETS) method to automatically adjust proper exposure time of each lens for different scenes. It guarantees at least two valid reference values for HDR image processing. Adopting the proposed weighting function restrains random distributed noise caused by micro-lens and produces a high quality HDR image. HDR image should be compressed to LDR image for being displayed on monitors, which is called tone mapping. This thesis proposes an integrated tone mapping methodology, which keeps all details in bright and dark parts. We use four cameras to implement these ideas. We first align these images to the same plane, and then adopt the methods we propose. The result image has extended dynamic range, that is, comprehensive information is provided. Eventually, we implement the entire system on Adlink MXC-6300 (Industrial Computer) platform that can reach 10 frames per second.