臺灣博碩士論文加值系統

隨著電子3C產品的價格降低，監控攝影系統的普及率也越來越高，在十字路口或停車場大多都有裝設攝影機，但目前攝影機僅具備記錄影像的功能，並不能發出警告或是判斷異常現象，為了能提升居住安全，故我們便針對一般的監控系統加入能偵測可疑車輛的偵測與路線記錄的功能。

本論文演算法採取的前景提取方法為背景相減，演算法分為兩大部分，一為多物件動態偵測，二為背景模型更新；多物件動態偵測的主要流程為利用切割影像畫面成多個搜索窗，只需判搜索窗(Search Window)內的前景資訊量是否足夠，再針對該Search Window來不斷位移尋找移動物體，省去對整張畫面做處理的運算時間與效能，而每一個Search Window即代表一個物體，當Search Window完整包覆物體時再利用統計量精確地鎖定物體的大小範圍，如此便能快速的使用搜索窗來區分多個物體；背景模型更新結合了背景相減法(Background subtraction)以及連續影像差異法(Temporal Difference)的效果達到了適應動態還境，根據偵測到的物體座標位置與停留時間來判斷是否融入或是脫離背景，來執行不同的背景更新效果。

硬體實作根據多物件偵測功能做硬體加速平行化運算，規律的排序記憶體儲存的資料，使讀取資料時更加有效率，達到即時運算物件偵測功能，硬體架構主要採用CIC的Cell-Based Flow和TSMC.90nm製程標準元件庫操作，每秒30張的畫面速率。

With the price of electronic 3C products reducing, the surveillance systems are more widely used at the crossroad and parking lots. However, the normal monitor only could record video. They cannot immediately aware people or analyze abnormal condition. For improving quantity of security, we add a new detect and record moving locus ability in normal surveillance system.

Based on our algorithm, we get the foreground data by using the background subtraction. There are two parts in this algorithm, including multi-objects motion detection and background update. In multi-object motion detection, first we spare many global search locations and judge if the number of foreground pixel is over the threshold. Second, shift the global search window and local search window to cover moving object. Third, we use the column and row histogram to get the object location and size. The global search and local search not only reduce the processing time and distinguish multi-object. In the background update, we mixed background subtraction and temporal difference to overcome dynamic background. We accord the moving object location and moving time to decide using the different update method.

To speed up software processing time, we implement the multi-object motion detection architecture in hardware design. Effectively manage memory data sorting improve processing speed. We use the cell-based design flow and TSMC 90 nm standard cell technology to process 30 frames per second.

第1章 序論 1
1. 1. 前言 1
1. 2. 研究動機 2
1. 3. 論文架構 4
1. 4. 實驗環境 5
第2章 相關研究 6
2. 1. 前景動態偵測 6
2. 1. 1. Background subtraction 6
2. 1. 2. Temporal difference 7
2. 1. 3. Optical flow 8
2. 2. 背景模式 9
2. 3. 移動估計演算法 10
2. 3. 1. 全域搜索演算法 11
2. 3. 2. 三步搜索演算法 12
2. 3. 3. 鑽石搜索演算法 13
第3章 移動偵測演算法與系統架構 14
3. 1. 系統流程簡介 14
3. 2. 色彩空間轉換 15
3. 2. 1. RGB Color model 15
3. 2. 2. YUV Color model 15
3. 2. 3. RGB轉灰階 16
3. 3. 背景相減 17
3. 3. 1. 二值化閥值設定 19
3. 4. 偵測移動物位置 22
3. 4. 1. 全域搜索 22
3. 4. 2. 區域搜索 27
3. 4. 3. 縮小鎖定範圍 28
3. 5. 背景更新 30
3. 5. 1. 偵測點分析 30
3. 5. 2. 背景更新模式 31
3. 6. 系統結果分析 41
3. 6. 1. 系統平台 41
3. 6. 2. 數據統計 41
3. 6. 3. 實驗結果 46
第4章 硬體架構設計 52
4. 1. 硬體架構 53
4. 2. 各單元架構 54
4. 2. 1. 記憶單元 54
4. 2. 2. 計數判斷單元 56
4. 2. 3. 位址控制單元 57
4. 2. 4. 位移搜索單元 59
第5章 硬體架構實作結果 60
5. 1. 數位IC設計流程 60
5. 2. 晶片規格及腳位 61
5. 3. 晶片實作結果 62
5. 3. 1. SYNTHESIS 62
5. 3. 2. Layout 63
第6章 總結與未來工作 64
6. 1. 結論 64
6. 2. 未來工作 64
參考文獻 65

[1]Y. L. Tian and A. Hampapur, "Robust Salient Motion Detection with Complex Background for Real-Time Video Surveillance," IEEE Application of Computer Vision, pp. 30-35, 2005.

[2]S. Gupte, O. Masoud, R. F. K. Martin and N. P. Papanikolopoulos, "Detection and classification of vehicles," IEEE Transactions on Intelligent Transportation Systems, vol. 3, no. 1, pp. 37-47, Mar 2002.

[3]N. Friedman, and S. Russell. "Image segmentation in video sequences: A probabilistic approach," 13th conference on Uncertainty in artificial intelligence. Morgan Kaufmann Publishers Inc. pp. 175-181, 1997.

[4]C. Stauffer and W. E. L. Grimson, "Adaptive background mixture models for real-time tracking," IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 252, 1999.

[5]D. Gutchess, M. Trajkovics, E. Cohen-Solal, D. Lyons and A. K. Jain, "A background model initialization algorithm for video surveillance," IEEE International Conference on Computer Vision, vol. 1, pp. 733-740, 2001.

[6]A. Elgammal, D. Harwood and L. Davis, "Non-parametric model for background subtraction," Conference on Computer Vision, pp.751-767. 2000.

[7]A. Elgammal, R. Duraiswami, D. Harwood and L. S. Davis, "Background and foreground modeling using nonparametric kernel density estimation for visual surveillance," in Proceedings of the IEEE, vol. 90, no. 7, pp. 1151-1163, Jul 2002.

[8]P. KaewTraKulPong and R. Bowden. "An improved adaptive background mixture model for real-time tracking with shadow detection," Video-based surveillance systems, pp.135-144, 2002.

[9]B. K. Horn and B. G. Schunck, "Determining optical flow," Artificial intelligence, vol.17, pp.185-203, 1981.

[10]L. Wixson, "Detecting salient motion by accumulating directionally-consistent flow," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 774-780, Aug 2000.

[11]Y. S. Jehng, L. G. Chen and T. D. Chiueh. "An efficient and simple VLSI tree architecture for motion estimation algorithms," IEEE Transactions on Signal Processing, vol. 41, no. 2, pp. 889-900, Feb 1993.

[12]K . M. Yang, M. T. Sun and L. Wu, "A family of VLSI designs for the motion compensation block-matching algorithm," IEEE Transactions on Circuits and Systems, vol. 36, no. 10, pp. 1317-1325, Oct 1989.

[13]Y. K. Lai, "A memory efficient motion estimator for three step search block-matching algorithm," IEEE Transactions on Consumer Electronics, pp.644-651, 2001.

[14]D. S. Lee, "Effective Gaussian mixture learning for video background subtraction," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 5, pp. 827-832, May 2005.

[15]S. Apewokin, B. Valentine, L. Wills, S. Wills and A. Gentile, "Multimodal mean adaptive backgrounding for embedded real-time video surveillance," IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-6, 2007.

[16]K. Toyama, J. Krumm, B. Brumitt and B. Meyers, "Wallflower: Principles and practice of background maintenance," IEEE International Conference on Computer Vision, vol. 1, pp. 255-261, 1999.

[17]McFarlane, Nigel JB, and C. Paddy Schofield, "Segmentation and tracking of piglets in images," Machine vision and applications, vol. 8, pp.187-193, 1995.

[18]K Kim, K. Chalidabhongse, T. H. Harwood and L. Davis, "Real-time foreground–background segmentation using codebook model," Real-time imaging, vol. 11, pp. 172-185, 2005.

[19]Z. Sun, G. Bebis, and R. Miller, "On-road vehicle detection: A review," IEEE transactions on pattern analysis and machine intelligence, vol. 28, pp.694-711, 2006.

[20]N. A. Mandellos, I. Keramitsoglou and C. T. Kiranoudis, "A background subtraction algorithm for detecting and tracking vehicles," Expert Systems with Applications, pp. 1619-1631, 2011

[21]S. Brutzer, B. Höferlin and G. Heidemann, "Evaluation of background subtraction techniques for video surveillance," IEEE Conference on Computer Vision and Pattern Recognitio, pp. 1937-1944, 2011.

[22]R. Cucchiara, C. Grana, M. Piccardi and A. Prati, "Detecting moving objects, ghosts, and shadows in video streams," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 10, pp. 1337-1342, Oct. 2003.

[23]Mikic, Ivana, et al. "Moving shadow and object detection in traffic scenes," International Conference on Pattern Recognition, vol.1, pp. 321-324, 2000.

[24]S. Y. Chien, S. Y. Ma and L. G. Chen, "Efficient moving object segmentation algorithm using background registration technique," IEEE Transactions on Circuits and Systems for Video Technology, vol.12.7 pp. 577-586, 2002.

[25]R. Jain, R. Kasturi and B. G. Schunck, "Machine vision," vol. 5, New York: McGraw-Hill., 1995.

[26]M. Fathy and M. Y. Siyal, "An image detection technique based on morphological edge detection and background differencing for real-time traffic analysis," Pattern Recognition Letters, vol. 16.12, pp. 1321-1330, 1995.

[27]T. Meier and K. N. Ngan, "Video segmentation for content-based coding," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 9, no. 8, pp. 1190-1203, Dec 1999.

[28]Kim, Jong Bae, and Hang Joon Kim. "Efficient region-based motion segmentation for a video monitoring system," Pattern Recognition, vol.24, pp.113-128, 2003.

[29]T. Horprasert, D. Harwood and L. S. Davis, "A statistical approach for real-time robust background subtraction and shadow detection," IEEE International Conference on Computer Vision, vol. 99, 1999

[30]Y. L. Tian, M. Lu and A. Hampapur, "Robust and efficient foreground analysis for real-time video surveillance," IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 1182-1187, 2005.

[31]T. Koga, "Motion compensated interframe coding for video conferencing," Conference on Telecommunications, Nov 1981.

[32]J. Y. Tham, S. Ranganath, M. Ranganath and A. A. Kassim, "A novel unrestricted center-biased diamond search algorithm for block motion estimation," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 8, no. 4, pp. 369-377, Aug 1998.