臺灣博碩士論文加值系統

所謂知己之彼，百戰百勝，在現代化的戰爭中收集對方的情報是十分重要的一環，在開戰之前能預先獲知敵方的部署，就可以做出良好的防範與決策，而在現代的科技之中，衛星影像已經發展的十分健全，不只是因為衛星影像的解析度提高，加上衛星影像幾乎沒有死角，所以可以被廣泛的應用在軍事領域，因此我們提出一套在衛星影像中偵測車輛的方法，以利軍事上的部署與決策。
在衛星影像偵測標物上有許多的方法，而本篇所使用的方法為，利用衛星影像將其利用Canny所提出的邊緣偵測方式，找出其對應的edge map，再將原始影像經過一些前處理的步驟，過濾一些會影響偵測的不利因素，之後再配合一些處理，以找出可能是車輛的區域(ROI)，再將其結合edge map即可找出我們要的目標物。
而在我們實驗的結果中，證明了我們所提出的方法，在偵測衛星影像中的車輛是可行的，而且可以利用我們的方法，進一步的利用在偵測其他的目標物上。

It is very important to collect the information of enemy in modern war. If we can know how the enemy disposes their forces, then we can make a good decision and prevent the happening of worst situation. Recently, due to the fast and mature development of satellite technologies, the satellite scanning resolution has been uplifted higher. Moreover, the scanning range covering by satellites almost has no forbidden area. It can be easily applied for military purpose. In response to this need, a novel method is proposed to detect vehicles in satellite images.
There are many approaches presented to detect objects in satellite images. In this thesis, an edge-based approach is proposed using the Canny’s edge detection method to detect object’s edges in satellite images. Then, the related edge-map can be found from object’s edges. Simultaneously, original satellite images are also preprocessed to reduce some unwanted situations, and then use image processing techniques to find out the ROI (region of interest) which contains the vehicles. Finally, we can successfully detect all vehicles in satellite images from the edge-map together with the ROI image.
Experiments were conducted on various satellite images and the results show that our proposed method is feasible and effective in detecting vehicles presented in satellite images. Furthermore, we can utilize this approach to detect other objects in satellite images.

Abstract………………………………………………………………………………І
摘要………………………………………………………………………………… Ⅱ
目錄………………………………………………………………………………… Ⅲ
附圖目錄…………………………………………………………………………… Ⅳ
第一章 緒論…………………………………………………………………………1
1.1 研究動機……………………………………………………………………1
1.2 相關研究……………………………………………………………………1
1.3 系統架構……………………………………………………………………2
1.4 論文架構……………………………………………………………………3
第二章 前處理……………………………………………………………………… 5
2.1 Smoothing………………………………………………………………… 5
2.2 min/max Filters………………………………………………………… 7
2.2.1 Max Filter………………………………………………………… 7
2.2.2 Min Filter………………………………………………………… 9
2.2.3 影像相減………………………………………………………… 10
2.3 Auto-thresholding……………………………………………………… 11
2.4 Connected-Components………………………………………………… 15
2.5 直線逼近………………………………………………………………… 17
2.6 Morphology operations…………………………………………………19
2.6.1 膨脹…………………………………………………………………19
2.6.2 侵蝕…………………………………………………………………20
2.6.3 斷開…………………………………………………………………20
2.6.4 閉合…………………………………………………………………20
第三章 車輛偵測…………………………………………………………………22
3.1 系統概述…………………………………………………………………22
3.2 邊緣偵測…………………………………………………………………22
3.3 找出 ROI…………………………………………………………………27
3.4 車輛偵測…………………………………………………………………29
3.4.1 遮罩設計…………………………………………………………30
3.4.2 車輛偵測…………………………………………………………32
第四章 實驗結果…………………………………………………………………36
第五章 結論與未來工作…………………………………………………………39
5.1 結論………………………………………………………………………39
5.2 未來工作 …………………………………………………………………40
參考文獻……………………………………………………………………………41

[1] Arthur Filippidis, L.C. Jain,and N. Martin, “Fusion of intelligent Agents for the Detection of Aircraft in SAR images”, IEEE Transactions on Pattern Analysis and Machine intelligence, Vol.22, No. 4, April 2000.

[2] Byoung-Ki Jeon, Jeong-Hun Jang, and Ki-Sang Hong, “Road Detection in Spaceborne SAR Images Using a Genetic Algorithm”, IEEE Transactions on Geoscience and Remote Sensing, Vol. 40, No. 1, January 2002.

[3] Florence Tupin, Bijan Houshmand, and Mihai Datcu, “Road Detection in Dense Urban Areas Using SAR Imagery and Usefulness of Multiple Views”, IEEE Transactions on Geoscience and Remote Sensing, Vol. 40, No. 11, November 2002.

[4] John Canny, “A Computational Approach to Edge Detection”, IEEE Transactions on Pattern Analysis and Machine intelligence, Vol. PAMI-8, No. 6, November 1986.

[5] Karl Segl and Hermann Kaufmann, “Detection of Small Objects From High-Resolution Panchromatic Satellite Imagery Based on Supervised Image Segmentation”, IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 9, September 2001.



[6] Martino Pesaresi and Jon Atli Benediktsson, “A New Approach for the Morphological Segmentation of High-Resolution Satellite Imagery”, IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 2, February 2001.

[7] Tao Zhao and Ram Nevatia, “Car Detection in Low Resolution Aerial Image”, International Conference on Computer Vision, 2001.

[8] Yu Su, Ping S. Huang, Chi-Fang Lin and Te-Ming Tu, “A Target-Cluster Fusion Approach for Classifying High Resolution IKONOS Imagery”, 16th IPPR Conference on computer Vision, Graphics and Image Processing,2003.