- 學位論文
於使用動態視覺感測器的結構光系統中作投影圖像之編碼隱藏
Embedding Temporal Signal within Projected Images for Dynamic Vision Sensor-based Structured Light System
摘要
在立體感測領域中，結構光掃描被視為簡單而有效率的方法之一。 與其他技術相比如立體視覺、飛時攝影等，結構光可以提供較為準確 的物體深度資訊。然而，傳統上使用相機與投影機的結構光系統有低 掃描速率或易受環境光影響等缺點。此外，結構光的掃描圖形會使人 眼感到不舒服，進而限制這項技術的應用範圍。 在這篇論文中，我們設計並實作了一種新的結構光系統，此系統利 用動態視覺感測器取代傳統的照相機，藉此消弭上述如低速率與易受 環境光影響等由相機造成的缺點。另外，我們的研究設計了一種適用 於使用動態視覺感測器的結構光系統的時間訊號來取代傳統如顏色、 強度的編碼方式。此時間訊號可以隱藏於各種灰階圖形之中而不被人 眼所察覺。本系統的掃描速率可以達到100Hz 以上，依照實驗結果， 在1.5 公尺的距離內，我們系統的解碼正確率可以達到80% 以上。
並列摘要
Structured light systems are one of the most simple and effective tools to acquire 3D models. Among all methods in the field of 3D measurement, structured light provides the most accurate shape recovery compared to passive or physical techniques. However, the performance of conventional structured light is limited by practical constraints of the camera. Moreover, the patterns of the conventional structured are usually distractive to human eyes. In this thesis, we design and implement a novel structured light system that utilized an asynchronous vision sensor called DVS. By using DVS, our system can achieve high reconstruction rate and resistance to ambient light. The key innovation of our work is that we replace the conventional pattern codification method, i.e., colors, intensities,... with temporal symbols displayed at high frequency. Moreover, this temporal pattern can be modified to display any grayscale image for human eye. Therefore, the distractive patterns of conventional structured light system can be replaced by any image. The pattern rate of our system can reach over 100 Hz and the experiment result shows that the decode correct rate of our temporal signal is above 80% under the distance of 1.5m.
參考文獻
延伸閱讀
- 曾信淳（2010）。基於適應性資訊隱藏技術之光學彩色影像加解密系統研究〔碩士論文，亞洲大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0118-1511201215465800
- 蔡淵裕、林若筑、張嘉君（2018）。植基於影像內插與像素差值之可適應性高動態範圍影像資訊隱藏演算法。資訊安全通訊，24(2)，45-57。https://www.airitilibrary.com/Article/Detail?DocID=a0000270-201804-201804240015-201804240015-45-57
- 鄭雅欣（2011）。使用於視訊散佈來源追蹤之壓縮域數位浮水印設計〔碩士論文，國立中央大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0031-1903201314425976
- 洪璿凱（2011）。Moving Object Detection and Tracking Using Binocular Vision Based on Spatial Constraints of Static Environment〔碩士論文，淡江大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0002-1708201122121000
- 林丕軒（2010）。On Simulation and Experimental Verification of LED Angular and Planar Illumination Distribution〔碩士論文，崑山科技大學〕。華藝線上圖書館。https://doi.org/10.6828/KSU.2010.00018