洪一平臺灣大學:資訊網路與多媒體研究所林彥良Lin, Yen-LiangYen-LiangLin2010-05-052018-07-052010-05-052018-07-052009U0001-1708200901005100http://ntur.lib.ntu.edu.tw//handle/246246/180719在視訊安全監控的相關研究中車輛與行人偵測和顏色辨識是相當重要的議題。 在 論文中,我們將針對車輛與行人的偵測和顏色辯識技術進行探討。 在車輛行人 測方面,目前的方法大多使用2D資訊做為特徵,例如邊緣、顏色、輪廓、動 ...等。 其中只有少部分的研究使用3D的特徵。 本論文提出一套新的演算法使 3D資訊來偵測車輛和行人。 首先會使用背景模型的技術來取得前景移動物體, 於每一個前景移動物體,我們會利用相機的內外在參數來計算物體在3D空間中 大小。 我們使用calibration-free的方法來估測攝影機參數, 其方法只需要在場 點選長方體的六個點即可。 顏色辨識系統方面,我們會利用Bayesian分類器來 練所定義的顏色在HSV色彩空間的決策邊界, 然後依據車輛和行人影像中的像素 所定義的顏色區域的分布來決定其顏色。 經由實驗結果,所提出的方法都能有 的運作We propose a real-time intelligence surveillance system.Two important topics are studied, including vehicle and pedestrian detection, vehicle and pedestrian color classification. Existing pedestrian and vehicle detection algorithms utilize 2D cues of objects, such as pixel values, color and texture, shape information or motion. Some of them require heavy computation power and are thus prohibited from real-time applications. While many researchers focus on modeling objects based on 2D cues, the use of 3D cues in object detection are not well studied. In this paper we propose an algorithm that utilizes 3D cues to perform pedestrian and vehicle detection. The 3D cues of objects in a static scene monitored by a camera can be obtained using the intrinsic and extrinsic parameters of that camera. We apply a calibration-free method to estimate the camera parameters. This method simply requires users to specify 6 vertices on a cuboid in the scene. In the spect of vehicle color classification, we use Bayesian classifier to trained the decision boundaries of defined color in the HSV space, then determining the color of the object according to distribution of the the pixels in the vehicle and pedestrian images on the defined color region. Experiment results demonstrate our proposed method can work efficiently.Abstract vii ist of Figures xi ist of Tables xiii Introduction 1 .1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 .2 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Vehicles and Pedestrians Detection Using 3D Scales and 2D Shapes 3 .1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 .2 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 .3 Overview of the Proposed System . . . . . . . . . . . . . . . . . . . . . 7 .4 Moving Blob Detection in Video . . . . . . . . . . . . . . . . . . . . . . 8 .4.1 Background Modeling using Codebook Algorithm . . . . . . . . 9 .4.2 Shadow Removal . . . . . . . . . . . . . . . . . . . . . . . . . . 10 .5 3D Feature Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 .5.1 Camera Calibration using Cuboid Algorithm . . . . . . . . . . . 12 inhole Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 xuboid Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . 14 .5.2 2D Bounding Box Model . . . . . . . . . . . . . . . . . . . . . . 15 .5.3 3D Bounding Box Model . . . . . . . . . . . . . . . . . . . . . . 18 hape Kernel Database . . . . . . . . . . . . . . . . . . . . . . . 19 hamfer Matching . . . . . . . . . . . . . . . . . . . . . . . . . 20 D Bounding Box . . . . . . . . . . . . . . . . . . . . . . . . . 23 peed Up Template Matching Process . . . . . . . . . . . . . . . 24 .6 Verification Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 .7 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Vehicle and Pedestrian Color Classification Using Bayesian Classifier 29 .1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 .2 Related work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 .3 Overview of The Proposed System . . . . . . . . . . . . . . . . . . . . . 32 .4 Color Decision Boundary Using Bayesian Rule . . . . . . . . . . . . . . 33 .5 Color Classification Algorithm . . . . . . . . . . . . . . . . . . . . . . . 36 .6 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Conclusions and Future Work 41 .1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 .2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 ibliography 43application/pdf10165063 bytesapplication/pdfen-US車輛偵測行人偵測車輛色彩辨識行人色彩辨識vehicle detectionpedestrian detectionvehicle color classificationpedestrian color classification即時車輛和行人偵測與顏色辨識系統Real-Time Vehicle and Pedestrian Detection and Color Classification Systemthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/180719/1/ntu-98-R96944029-1.pdf