研究生: |
陳彥博 Yan-Bo Chen |
---|---|
論文名稱: |
基於深度資訊並使用局部形狀特徵之即時槍手辨識系統 Real Time Gunner Detection System Using LSC Based on Depth Image |
指導教授: |
洪西進
Shi-Jinn Horng |
口試委員: |
林韋宏
Wei-hong Lin 高宗萬 Tzung-Wan Gau 顏成安 Cheng-An Yen |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 資訊工程系 Department of Computer Science and Information Engineering |
論文出版年: | 2013 |
畢業學年度: | 101 |
語文別: | 中文 |
論文頁數: | 43 |
中文關鍵詞: | 槍手 、深度影像 、端點偵測 、分類器 |
外文關鍵詞: | gunmen, depth image, endpoint detection, classifier |
相關次數: | 點閱:479 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
槍枝的影像辨識與資訊安全的議題已被討論許久,若能有效的從監視畫面中自動偵測到持槍歹徒,對於社會的治安維護能夠有相當大的助益,也能保障一般市民的人身安全。
槍體本身含有的特徵紋理包含外型與體積兩類,而這兩種特徵在一般背景之下較為常見,其餘的特徵則皆不明顯,所以在一般背景很難辨識出槍的存在。
本論文提出一個槍手偵測的系統,使用深度資訊為基礎,能夠在不受其背景複雜度或光影變化影響的情況下,有效的過濾背景,從而取出前景物件。再以手部持槍的方式結合槍與手兩者的外型作為特徵分類的訓練,在辨識過程中先找出手部等端點部位,並利用之前訓練好的分類器辨識使否持槍,達到偵測槍手的目的。
The issues of guns image recognition and information security have been discussed for a long time. It will help the society security, that if it detects gunmen effectively from the monitor screen. And it can protect the personal safety of the general public.
Gun body contains characteristics of square shape and small, however these characteristics are more common than the remaining ones in the general context. Therefore it is difficult to discriminate whether the gunmen exist or not in the general context.
This thesis presents a gunman detection system. It uses the depth coding so that it cannot be affected by the background complexity or brightness variation. And it also can filter characteristics effectively in the background, thereby getting the foreground object. Then we train a classifier by using the features of gun-holding-hand to combine both appearance characteristics and discriminate whether armed or not in the parts of endpoints like hand endpoint.
[1] Zhenning Li and Kulic, D., “Local Shape Context Based Real-time Endpoint Body Part Detection and Identification from Depth Images,” Computer and Robot Vision (CRV) 2011, Page(s): 219 – 226
[2] Lienhart, R. and Maydt, J., “An Extended Set of Haar-Like Features for Rapid Object Detection,” Image Processing 2002, Page(s): I-900 - I-903 vol.1
[3] Young-Shin Lim, Hyonam Joo and Joon-Seek Kim, “A real-time pattern detection using fast feature matching algorithm,” ICCAS 2008. Page(s): 1981 – 1985
[4] Hui Zeng, Ji-Yuan Dong, Zhi-Chun Mu and Yin Guo, “Ear Recognition Based on 3D Keypoint Matching,” ICSP, IEEE 2010 , Page(s): 1694 – 1697
[5] Xianchun Zheng, Kamata S. and Liang Yu, “Face recognition with Local Gradient Derivative Patterns,” TENCON IEEE 2010 , Page(s): 667 - 670
[6] Sun Junding, Zhu Shisong and Wu Xiaosheng, “Image Retrieval Based on an Improved CS-LBP Descriptor,” ICIME, IEEE 2010 , Page(s): 115 - 117
[7] Cheng-Sheng Chang, “Rapid Gunman detection for Surveillance System Using Motion Information and Template Matching Technique,” 2010
[8] Kwan-Jung Oh, Vetro A. and Yo-Sung Ho, “Depth Coding Using a Boundary Reconstruction Filter for 3-D Video Systems,” 2011 , Page(s): 350 – 359
[9] Zheng Yi and Fan Liangzhong, “Moving object detection based on running average background and temporal difference,” ISKE 2010, Page(s): 270 – 272
[10] Chia-Hsiang Chen, “An Intelligent Surveillance Based on Statistical Gaussian Mixture Background Model with Kalman Filter,” 2010
[11] Shuo Chen, Cheng-dong Wu and Xiao-sheng Yu, “Fast scene recognition based on saliency region and SURF,” ICICIP 2011, Page(s): 863 – 866
[12] Jiaxiang Huang, Shaozi Li and Changle Zhou, “Extension of Mean Shift vector with theoretical analysis and experiment,” ISKE 2008, Page(s): 1007 – 1012
[13] Freedman, D. and Kisilev P., “Fast Mean Shift by compact density representation,” CVPR 2009, Page(s): 1818 – 1825
[14] Yih-chuan Lin and Shen-Chuan Tai, “Fast full-search block-matching algorithm for motion-compensated video compression,” 1997, Page(s): 527 – 531
[15] Brunig M. and Niehsen W., “Fast full-search block matching,” 2001, Page(s): 241 – 247
[16] Seyedhosseini M., Paiva A., R.C. and Tasdizen T., “Fast AdaBoost training using weighted novelty selection,” IJCNN 2011, Page(s): 1245 – 1250
[17] Donglai Xu, Bailey C. and Sotudeh R., “An improved three-step search block-matching algorithm for low bit-rate video coding applications,” ISSSE 98. 1998, Page(s): 178 – 181
[18] Seth K., Rangarajan P., Srinivasan S., Kamakoti V. and Bala Kuteshwar V., “A parallel architectural implementation of the New Three-Step Search algorithm for block motion estimation,” ICVD 2004, Page(s): 1071 – 1076
[19] Khongpit V., Phangphol U., Kasemsan K. and Srisa-an C., “2D Gun's type classification using edge detection algorithm and SUSAN low level image processing,” 2012, Page(s): 521 - 523
[20] 警政統計通報(102年第11號)
[21] Li Chengdong, Li Ming and Du Bianjing, “AdaBoost algorithm on face detection,” 2011
[22] “Hierarchical clustering,” http://mirlab.org/jang/books/dcpr/dcHierClustering.asp