外文翻譯---一個魯棒的基于機器視覺運動目標檢測與跟蹤系統(tǒng)

1、附錄 附錄 I 外文文獻翻譯 外文文獻翻譯(1)原文： 原文：A Robust Vision-based Moving Target Detection and Tracking System AbstractIn this paper we present a new algorithm for real~time detection and tracking of moving targets in terrestrial scen

2、es using a mobile camera. Our algorithm consists of two modes: detection and tracking. In the detection mode, background motion is estimated and compensated using an affine transformation. The resultant motion rectifie

3、d image is used for detection of the target location using split and merge algorithm. We also checked other features for precise detection of the target location. When the target is identified, algorithm switches to the

4、tracking mode. Modified Moravec operator is applied to the target to identify feature points. The feature points are matched with points in the region of interest in the current frame. The corresponding points are furthe

5、r refined using disparity vectors. The tracking system is capable of target shape recovery and therefore it can successfully track targets with varying distance from camera or while the camera is zooming. Local and regio

6、nal computations have made the algorithm suitable for real-time applications. The refined points define the new position of the target in the current frame. Experimental results have shown that the algorithm is reliable

7、and can successfully detect and track targets in most cases. Key words: real time moving target tracking and detection, feature matching, affine transformation, vehicle tracking, mobile camera image.1 Introduction Visual

8、 detection and tracking is one of the most challenging issues in computer vision. Application of the visual detection and tracking are numerous and they span a wide range of applications including surveillance system, ve

9、hicle tracking and aerospace application, to name a few. Detection and tracking of abstract targets (e.g. vehicles in general) is a very complex problem and demands sophisticated solutions using conventional pattern reco

10、gnition and motion estimation methods. Motion-based features as well as constraint on area of the target as discussed in this section.2.1 Background motion estimationAffine transformation [8] has been used to model motio

11、n of the camera. This model includes rotation, scaling and translation. 2~D affine transformation is described as follow: (1) ? ? ??? ? ?? ? ? ? ??? ? ??? ??? ?? ?? ? ??? ? ??a ay xa aa aY Xiiii654 32 1where (xi , yi ) a

12、re locations of points in the previous frame and (Xi , Yi ) are locations of points in the current frame and a1~a6 are motion parameters. This transformation has six parameters; therefore, three matching pairs are requir

13、ed to fully recover the motion. It is necessary to select the three points from the stationary back~ground to assure an accurate model for camera motion. We used Moravec operator [9] to find distinguished feature points

14、to ensure precise match. Moravec operator selects pixels with the maximum directional gradient in the min~max sense. If the moving targets constitute a small area (i.e. less than 50%) of the image, then LMedS algorithm c

15、an be applied to determine the affine transformation parameters of the apparent background motion between two consecutive frames according to the following procedure. 1. Select N random feature point from previous frame,

16、 and use the standard normalized cross correlation method to locate the corresponding points in the current frame. Normalized correlation equation is given by: (2)2 1, ,22 221 1, 2 2 1 1] ) , ( [ ] ) , ( [] ) , ( ][ ) ,

17、( [? ?? ? ?? ? ? ? ?? ??? ??? ??S y x S y xS y xf y x y xy x y xrf f ff f f fhere and are the average intensities of the pixels in the two regions being 1 f 2 fcompared, and the summations are carried out over all pixe

18、ls with in small windows centered on the feature points. The value r in the above equation measures the similarity between two regions and is between 1 and -1. Since it is assumed that moving objects are less than 50% of