(18.206.238.77) 您好!臺灣時間:2021/05/17 18:58
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:趙晉煌
研究生(外文):Jin-Huang Jaho
論文名稱:應用正規化色彩與特徵分解法於車輛偵測
論文名稱(外文):Application of Normalized Color and Feature Decomposition Method to Vehicle Detection
指導教授:張名輝
指導教授(外文):Ming-Hui Chang
口試委員:陳沛仲陳炘鏞陳世宏楊錫凱張名輝
口試委員(外文):Pei-Chung ChenShin-Yong ChenShih-Hung ChenShyi-Kae YangMing-Hui Chang
口試日期:2011-07-22
學位類別:碩士
校院名稱:遠東科技大學
系所名稱:機械工程研究所
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2011
畢業學年度:99
語文別:中文
論文頁數:78
中文關鍵詞:車輛偵測正規化色彩特徵分解法類神經網路
外文關鍵詞:vehicle detectionnormalized colorfeature decompositionneural network
相關次數:
  • 被引用被引用:2
  • 點閱點閱:174
  • 評分評分:
  • 下載下載:20
  • 收藏至我的研究室書目清單書目收藏:1
本論文提出一個以視覺影像為基礎,應用正規化色彩與特徵分解法於車輛偵測。以靜態圖片作車輛偵測,在固定的場景裡面,找出車輛並且作車輛定位,可應用在停車場、高速公路與停車違規上。主要方法分為三個階段:
第一個階段:使用一種色彩空間轉換方法,就像人臉偵測時會先尋找膚色區域一般,快速地找出影像中屬於疑似車輛區域。
第二個階段:在找出疑似屬於車輛之區域後,使用特徵分解法,在車輛各種不同的方向(直、橫、左斜、右斜),計算每塊目標圖像在水平、垂直方向的差異值。
第三個階段:訓練車輛樣本,將各種方向的車輛特徵,送入類神經網路,經由學習法則調整權重值,讓權重值中包含各種方向的車輛變化,最後所得到的權重值,與差異值資料做比對,比對成功在車輛位置畫上方框,此方框的位置即代表所欲偵測的車輛。
在所測試的81張測試圖片中,完全辨識正確的車輛有304台,無法辨識為車輛的有42台,誤判其他區域為車輛有19台,辨識率為87.86%,誤判率為5.49%。
關鍵字:車輛偵測、正規化色彩、特徵分解法、類神經網路。
In this study, a static image method for vehicle detection is proposed. In the stationary scenario, the proposed method recognizes the vehicle target and finds the location in the image. The application of proposed method can be used in the parking lot management, highway traffic control and parking violations expositions. This approach uses normalized color and feature decomposition method to vehicle detection based on visual images. The proposed method can be divided into three stages:
The first stage uses the color space transformation method (normalized color method). As the human face detection method, a quick way to location the human position is to find the skin area firstly. In this stage, a normalized color space transformation method would be introduced to quickly identify suspected vehicles location in images.
The second stage uses feature decomposition method in the area obtained by the first stage. The feature decomposing method would calculate the deviations of horizontal and vertical direction of suspected area. These data would be used in third stage as the vehicles identification patterns.
The third stage is a training process. The vehicle feature patterns with different direction obtained by previous stage would be the training samples. In this study, a neural network would be used as the classify mechanism to distinguish the vehicle area and non-vehicle area. Finally, the output of neural network would be the candidates of the vehicles. After some image process to eliminate the false candidates, the position of vehicles in images could be obtained.
In all the 81 test images, there are 304 vehicles to be exactly correct detection. However, there are 42 vehicles cannot be recognized correctly and 19 objects to be false alarm. Finally, the recognition rate of proposed method is 87.86%, the false positive rate is 5.49%.
Keyword: vehicle detection, normalized color, feature decomposition, neural network.
誌 謝 .................................................................................................................. i
摘 要 ................................................................................................................. ii
目 錄 ................................................................................................................ iv
表目錄 ............................................................................................................. vii
圖目錄 ............................................................................................................ viii
第一章 緒論 ..................................................................................................... 1
1.1 研究背景 .......................................................................................... 1
1.2 研究動機 .......................................................................................... 2
1.3 研究方向與目的 .............................................................................. 3
1.4 研究流程與架構 .............................................................................. 4
第二章 文獻回顧與探討 ................................................................................. 7
2.1 動態資訊檢測法則 .......................................................................... 7
2.2 靜態資訊檢測法則 .......................................................................... 9
2.2.1 車輛顏色 ............................................................................... 9
2.2.2 車輛特徵 ............................................................................. 11
2.2.3 背景資料 ............................................................................. 13
第三章 影像處理理論 ................................................................................... 14
3.1 車輛顏色探測器 ............................................................................ 14
3.1.1 顏色特徵的降維方法 ......................................................... 14
3.1.2 應用貝氏分類器於像素分類 ............................................. 19
3.2 特徵分解法 .................................................................................... 20
3.3 類神經網路 .................................................................................... 23
3.3.1 類神經網路基本架構 ......................................................... 24
3.3.2 倒傳遞類神經網路(Back-propagation Nenral Network) ... 26
第四章 車輛偵測法 ....................................................................................... 30
4.1 正規化色彩 .................................................................................... 31
4.2 圖塊分類檢驗 ................................................................................ 34
4.3 八連通標示法 ................................................................................ 38
4.4 改良式特徵分解 ............................................................................ 42
4.5 倒傳遞類神經網路學習與回想 .................................................... 44
4.6 框數融合法 .................................................................................... 51
4.7 強化法 ............................................................................................ 55
4.7.1 正規化色彩改良 ................................................................. 55
4.7.2 特徵分解法補強 ................................................................. 57
4.7.3 區塊樣本篩檢 ..................................................................... 57
第五章 實驗結果與討論 ............................................................................... 62
5.1 硬體架構 ........................................................................................ 62
5.2 結果與討論 .................................................................................... 63
5.3 辨識速度測試 ................................................................................ 69
第六章 結論與未來發展 ............................................................................... 70
6.1 結論 ................................................................................................ 70
6.2 未來發展 ........................................................................................ 73
參考文獻 ......................................................................................................... 74
自 述 ............................................................................................................... 78
[1]. 林大傑,“車輛偵測器不完整資訊推估旅行時間之研究”,逢甲大學交通工程與管理學系碩士論文, July,2007。
[2]. A. Bensrhair, M. Bertozzi, A. Broggi, P. Mich´e, S. Mousset, and G.Toulminet, “A Cooperative Approach to Vision-Based Vehicle Detection” , IEEE Intelligent Transportation Systems Conference Proceedings, August, 2001, pp.25-29.
[3]. Surendra Gupte, Osama Masoud, Robert F. K. Martin, and Nikolaos P. Papanikolopoulos, “Detection and Classification of Vehicles” , IEEE Transactions On Intelligent Transsportation Systems, 2002.
[4]. Zhenfeng Zhu, Hanqing Lu, James Hu, Keiichi Uchimura, “Car Detection Based on Multi-Cues Integration” , Pattern Recognition, 2004.
[5]. Alberto Broggi, Pietro Cem, Pier Claudio Antonello, “Multi-Resolution Vehicle Detection Using Artificial Vision”, IEEE Intelligent Vehicles Symposium University of Parma Rarma, June, 2004, pp.14-17.
[6]. 張仁鴻,“智慧型監視系統之物體偵測與交通影片之車輛偵測與索引”,國立雲林科技大學電機工程系碩士班,July,2004。
[7]. Guolin Wang, Deyun Xiao, Gu J, “Review on Vehicle Detection Based On Video for Traffic Surveillance”, Automation and Logistics, 2008.
[8]. Wen Chung Chang, “Online Boosting for Vehicle Detection”, IEEE Transactions On Systems, Vol. 40, No. 3, June, 2010.
[9]. Xinting Pan,Yunlong Guo, Aidong Men, “Traffic Surveillance System for Vehicle Flow Detection”, Computer Modeling and Simulation, 2010.
[10]. Jinhui Lan, Meng Zhang, “A New Vehicle Detection Algorithm for Real-time Image Processing System”, Computer Application and System Modeling (ICCASM), 2010.
[11]. Guangtao Cheng,Xue Chen, “A Vehicle Detection Approach Based on Multi-Features Fusion in the Fisheye Images”, Computer Research and Development (ICCRD), 2011.
[12]. Luo-Wei Tsai, Jun-Wei Hsieh, “Vehicle Detection Using Normalized Color and Edge Map”, IEEE Transactions on Image Processing, Vol. 16, NO. 3, March, 2007.
[13]. Nan Zeng and Jill D. Crisman, “Vehicle Matching Using Color”, IEEE Intelligent Transportation System, 1997, pp. 206-211.
[14]. S. Romani, P. Sobrevilla, E. Montseny, “On the Reliability Degree of Hue and Saturation Values of a Pixel for Color Image Classification”, IEEE International Conference on Fuzzy Systems, 2005.
[15]. Ku-Jin Kim, Sun-Mi Park, Yoo-Joo Choi, “Deciding the Number of Color Histogram Bins for Vehicle Color Recognition”, IEEE Asia-Pacific Services Computing Conference, 2008.
[16]. Xiuzhi Li, Guangming Zhang, Jing Fang, Jian Wu, Zhiming Cui, “Vehicle Color Recognition Using Vector Matching of Template”, Third International Symposium on Electronic Commerce and Security, 2010.
[17]. Jing Fang, Hengjun Yue, Xiuzhi Li, Jian Wu Zhiming Cui, “Color Identifying of Vehicles Based on Color Container and BP Network”, Business Management and Electronic Information (BMEI), 2011.
[18]. Lin-Cheng Wang, Syed A. Rizvi,and Nasser M. Nasrabadi, “Automatic Target Recognition Using a Feature-Decomposition and Data-Decomposition Modular Neural Network”, IEEE Transactions on Image Processing, Vol. 7, NO. 8, August , 1998.
[19]. 賴韋誠,“一種簡單的邊緣影像特徵擷取法”,元智大學資訊工程研究所碩士論文,June,2002。
[20]. Zehang Sun, George Bebis, and Ronald Miller, “Monocular Precrash Vehicle Detection: Features and Classifiers”, IEEE Transactions on Image Processing, Vol. 15, NO. 7, July, 2006.
[21]. Yong Chuan, Chong Qing, China, “A Study on Contour Feature Algori thm for Vehicle Type Recognition”, Artificial Intelligence, 2009.
[22]. Junwen Wu, Xuegong Zhang and Jie Zhou, “Vehicle Detection in Static Road Images with PCA-and-Wavelet-Based Classifier“, IEEE Intelligent Transportation Systems Conference Proceedings, August, 2001.
[23]. Xuezhi Wen, Hong Zhao, Nan Wang, Huai Yuan, “A Rear-Vehicle Detection System for Static ImagesBased on Monocular Vision”, Control, Automation, Robotics and Vision, 2006.
[24]. 鄭欽任,“以線段特徵為基礎的靜態影像車輛偵測”,國立臺灣科技大學電機工程系碩士論文,July,2006。
[25]. R.Aarthi, S.Padmavathi, J.Amudha, “Vehicle Detection in Static Images Using Color and Corner Map”, Recent Trends in Information, Telecommunication and Computing (ITC), 2010.
[26]. Z. Sun, G. Bebis, and R. Miller, “On-road Vehicle Detection Using Optical Sensors:A View,” IEEE Intelligent Transportation Systems Conference, Washington, D.C., USA, Oct. 3-6, 2004, pp.585-590.
[27]. Glenn Healey, “Segmenting Images Using Normalized Color”, IEEE Transactions On Systems, Vol 22. NO 1. January / February,1992.
[28]. J. C. Rojas and J. D. Crisman, “Vehicle Detection in Color Images,” IEEE Conference on Intelligent Transportation System, Nov.9-11,1997, pp.403-408.
[29]. 蔡洛緯,“以正規化色彩與邊緣資訊作車輛偵測”,國立中央大學資訊工程研究所碩士論文,June,2005。
[30]. 施柏屹,“倒傳遞類神經網路學習收斂之初步探討”,國立中央大學機械工程研究所碩士論文,July,2000。
[31]. 王進德,蕭大全,“類神經網路與模糊控制理論入門”,全華科技圖書,初版,2001。
[32]. 張斐章、張麗秋,“類神經網路”,東華書局,初版,2005。
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
1. 蕭婉鎔、黃同圳,2009,〈策略性人力資源管理理論之驗證-普遍性、構形及權變觀點〉,《管理與系統》,16(1):1-23。
2. 蕭欣瑋,2008,〈我國公部門臨時人員進用及管理之研究〉,《人事月刊》,46(6):2-8。
3. 謝正君,2003,〈契約式聘僱人力之探討- 以台北市政府為例〉,《公務人員月刊》,81:27-33。
4. 蔡秀涓,2007,〈全球與在地之趨同與差異:台灣公部門策略性人力資源管理初探〉,《人事月刊》,45(3):38-43。
5. 蔡良文,2008,〈政府契約用人制度變革之研析〉,《考銓季刊》,53:63-94。
6. 葉德豐、劉紫娟、柯瑞芬、陳進堂、王俊文等,2004,〈公立醫院非醫師正式人員與臨時約聘僱人員之激勵態度、工作滿意度與離職傾向研究-以某公立區域醫院為例〉,《醫護科技學刊》,6(3):243-258。
7. 曾介宏,2006,〈公部門推動策略性人力資源發展的基本問題〉,《研習論壇月刊》,72 :30-38。
8. 張火燦,1998,《策略性人力資源管理》(第二版),台北市:揚智文化事業股份有限公司。
9. 陳金貴,2005,〈政府改造對公部門人力資源影響管理之探討〉,《公務人員月刊》,106:15-21。
10. 許南雄,1998,〈各國政府組織員額精簡概說〉,《人事月刊》,27(6):29-35。
11. 孫本初,2007,〈策略性人力資源管理在政府機關運用與行銷〉,《人事月刊》,44(1):4-11。
12. 施能傑、蔡秀涓,2004,〈契約性人力制度之規劃作法〉,《公務人員月刊》,102:15-34。
13. 施能傑、蔡秀涓,2003,〈契約性人力運用之理論與現實〉,《公務人員月刊》,81:15-26。
14. 施能傑,2001,〈彈性化職位設計與政府人力運用〉,《人事月刊》,32(5):33-45。
15. 邱忠民,2008,〈行政核心價值及首長領導型態對地方公務人員離職流動傾向之影響研究-以台南縣政府為例〉,《嘉南學報》,34:780-790。