使用二維彩色圖像修補紅外線深度圖像

The Needs for Advance Care Planning in Cancer Patients

現今流行的深度成像方法包括結構光（structure light）、飛行時間（time of light）和視差立體圖像（stereoscopic image）。由於任何點雲檢測方法皆可能產生深度資訊遺失，因此有必要發展深度影像修補方法。由於現今三維彩色圖像的採集皆採用二維彩色圖像與深度圖像相結合，因此，我們提出了一種使用二維彩色圖像修補紅外線深度圖像之修補程序，並提出使用迭代低通擴張法進行深度資訊修補，具有通用且簡單之特性。我們在實驗中產生較強的紅外雜訊來破壞部分深度圖像，然後利用新提出之方法成功地對深度圖像進行了修補。本研究亦建立模擬分析流程，證明了迭代低通擴張法具有極好的準確度和精確度。 The most popular depth imaging methods including structure light, time-of-flight, and stereoscopic image. Since no point cloud detection method can prevent depth information loss, the complement methods is therefore important. Conventional 3D-image technologies produce RGBD images use 2D RGB image combing with depth image. Therefore, we proposed a depth information complement method that repair the IR-depth image with use of 2D RGB image. Besides, we propose a novel iterative low-pass pervasion method for depth image pervasion. In the experiment, we use strong IR noise to destroy parts of the depth image. Then we successfully complement the depth image by the proposed method. This research also builds a simulation analyzing process. It demonstrates that iterative low-pass pervasion method has good accuracy and precision.