主動式雷射尋標器前瞻技術研究

Abstract

隨著雷射測距儀的不斷發展，現今已應用在許多不同的領域，像是地形地貌監測、 三維立體物的量測等等。藉由三維雷射影像掃描儀能取得物體的距離影像以及強度影 像，可以提供物體表面的三度空間的資訊。相較於傳統利用立體視覺影像產生的深度資 訊，三維雷射影像掃描儀除了有不需要校正相機參數的優點外，還能提供較精確的量測 結果。利用這些特點，可以在取得物體表面資訊後，發展一套三次元影像識別的技術， 進行三維物體的辨識。 本計畫將建立雷射尋標之前瞻技術的基礎，目標規格為≧100 kHz，平均輸出功率 ≧5 W，脈衝寬度50-100 ns 的高重複率光纖雷射。藉由本計畫建立雷射尋標之前瞻技術 的研究基礎，並朝著脈衝能量、脈衝寬度符合軍方需求的高重複率光纖雷射努力。未來， 如何強化雷射的穩定與耐衝擊性，包含如何將系統光纖化等關鍵技術，也是值得研究的 方向。本計畫另一重點係針對三次元影像辨識技術的發展進行研究，主要可以分為五大 部份：(a)場景點雲資料的前處理，包含雜訊濾除與背景剔除 (b)點雲資料的三維幾何特 性分群，將點雲資料根據不同的三維幾何特性加以分類，應用於場景的分割 (c)以三維 線索法為基礎的快速資料庫檢索，找出分割後的區域場景所對應的可能目標物模型 (d) 多重精細度的點雲資料處理，使得在任意解析度下取得的場景，都可以對應到合適的目 標物模型 (e)建立三維表面特徵與研發識別法則，使用一個與姿態無關的物體特徵，可 以解決物體的特徵影像不隨著姿態改變而改變。最後，整合以上五個部份，發展出本計 畫的三維物體偵測與識別法則。

The laser range finder technology is used in many domains, like distance measurement, terrain monitor and 3-D object measurement. Both the range image and the intensity image of object will be acquired by the 3-D image laser scanner. The range image can provide 3-D information about object. Compares with depth information which is produced by the traditional stereoscopic vision image, the advantages of using 3-D image laser scanner are avoiding camera calibration and obtaining accurate result. Hence, the 3-D surface contour is obtained by 3-D image laser scanner can be used for 3-D object recognition. In order to develop an algorithm of 3-D object recognition, the most important thing in this project is to extract a pose-independent feature. High-power light sources in the eye-safe wavelength regime near 1.55-m have many potential applications in free-space communication, range finding, nonlinear frequency conversion, and medical surgery. These prospective applications have led to the erbium-ytterbium-codoped double-clad fiber laser to receive a great deal of interest during the past years. Er3+ and Yb3+ codoped gain materials are generally significantly superior to Er3+-doped gain media because the incorporation of Yb3+ ions considerably enhances the pump absorption, reduces the clustering of Er3+ ions, and allows for higher Er3+ concentrations without strong quenching effects. Q Switching is an effective approach to intensify the laser peak power. Among various Q switching methods, passive Q switching by use of a saturable absorber is a simple, convenient, and efficient way to achieve high-peak-power pulses. However, the actively Q-switched fiber lasers generate more stable and controllable pulse train. For specific military purpose, the target of this project is to develop the 5-W eye-safe actively Q-switched fiber laser with the pulse width of 50-100 ns and the repetition rate of the pulse laser would be more than 100 kHz. The study of this 3-D object recognition project divides into five parts: (a) the pre-processing of point cloud data (PCD), including 3D noise filtering and background elimination, (b) a point cloud classification based on the type of surface the point is lying on, (c) a fast model retrieval method based on 3D cueing, this can be applied to scene segmentation and finding the candidate models in the database, (d) multi-resolution point cloud processing, this approach handles the consistency of the sampling data in different resolutions, and (e) creating surface features and developing 3-D object recognition algorithm, the concept of spin image is introduced in pose-independent feature extraction. In short, all pose-independent features of 3-D models in this project are generated based on this characteristic. Indeed, the 3-D object recognition approach integrates all above parts in this project.