臺灣博碩士論文加值系統

在本篇論文中,我們設計了一個電腦視覺追蹤系統。整個系統可以分為兩個主要部份,第一個部份為在影像中檢測出目標物的位置,第二個部份則為目標物軌跡的估測。
我們以目標物的邊緣分佈為基礎來偵測目標物在畫面中的位置。首先目標物的邊緣分佈會從目標物模型中求得,再將畫面中所有的邊緣分佈求出,最後將兩者做旋積分,以求出目標物所在的位置。這種方法應用在複雜背景時比傳統以灰階強度為基礎的模版比對法有更高的可靠度。
在求出目標物在畫面中的位置之後,一個以交互作用多模型演算法為基礎的軌跡追蹤方法被應用在目標物的追蹤及相機平台的控制。藉由這個方法可以追蹤目標物相對於相機平台的角速度及角加速度。
最後,一些實驗將證明我們的視覺追蹤系統應用在複雜背景的追蹤能力。

In this thesis, a visual tracking system is proposed. The whole system is divided into two parts. The first part is the detection of the target, and the second part is the trajectory filtering.
To detect the target in the image, we use the method based on the distribution of edges. In this method, the template is edge based, and the distribution of edges of the image is extracted, so that the two are convolved to find the location of the target. The method is more reliable than the traditional template matching in the complicated scene.
After finding the target in the image, we will control the camera platform to keep the target about the center of the image. The location of the target in the world coordinate frame is first transformed to the orientation relative to the camera. Where by a tracking algorithm based on the IMM (Interacting Multiple Model) algorithm is developed. The algorithm estimates the angular velocity and the angularity acceleration of the target relative to the camera, which along with some error information in the images are used to control the camera platform.
Finally, a number of experiments are conducted to validate our visual tracking system for tracking the targets in a complicated scene.

Chapter1. Introduction
1.1 Motivations
1.2 Problem Definition
1.3 Thesis Organization
Chapter2. Preliminaries
2.1 System Overview
2.2 Coordinate Transformation of the System
2.3 Review of Target Detection Methods
2.4 Discriminative Signal to Noise Ratio
Chapter3. Target Detection through Images
3.1 Template Matching with SAD
3.2 Matched Spatial Filter
3.3 Template Matching Based on Edge Distribution
3.4 Implementation of the Matching Process
3.5 Advantages and Limitations
Chapter4. Control of Camera Platform
4.1 Interacting Multiple Model
4.2 Design of Tracking Algorithm
4.3 Simulations
Chapter5. Experiments
5.1 Smooth Motion
5.2 Abrupt Motion
5.3 Free Motion
Chapter6. Conclusions and Future Works
References

[1]. DAE-SIK JANG, GYE-YOUNG KIM and HYUNG-IL CHOI, "Model-based Tracking of Moving object", Pattern Recognition, Vol. 30, No.6, pp.999-1008, 1997.
[2]. R. Jain, "Extraction of motion information from peripheral processes", IEEE Trans. Pattern Analysis Mach. Intell. PAMI-3, pp.489-503, 1981.
[3]. Han Wang, Choon Seng Chua and Ching Tong Sim, "Real-time Object Tracking from Corners", Robotica, Vol. 16, pp.109-116, 1998.
[4]. M. Kass, A. Witkin and D. Terzopoulos, "Snakes: active contour models", Proc. 1st. int. Conference on Computer Vision, London, pp.259-268, 1987.
[5] Jezekiel Ben-Arie and k. Raghunath Rao, "Optimal Template Matching by Nonorthogonal Image Expansion Using Restoration", Machine Vision and Applications, Vol. 7, pp.69-81, 1994.
[6] R. Brunelli and T. Poggio, "Template Matching: Matched Filter and Beyond", Pattern Recognition, Vol. 30, No.5, pp.751-768, 1997.
[7] B. S. Manjunath, C. Shekhar and R. Chellappa, "A New Approach to Image Feature Detection with Applications", Pattern Recognition, Vol. 29, No. 4, pp.627-640, 1996.
[8] Anil K. Jain, yu Zhong and Sridhar Lakshmanan, "Object Matching Using Deformable Templates", IEEE Trans. on Pattern Analysis and Mach. Intell., Vol. 18, No. 3, March 1996.
[9] J.L. Edwards and H.Murase, "Coarse-to-fine Adaptive Masks for Apperance Matching of occluded Scenes", Machine Vision and Applications, Vol.10, pp.232-242. 1998.
[10] Chun-Hsien and Yung-Chang Chen, "Moment Preserving Pattern Matching", Pattern Recognition, Vol. 23, No. 5, pp. 461-474, 1990.
[11] E. Mazor, A. Averbuch, Y. Bar-Shalom and J. Dayan, "Interacting Multiple Models in Target Tracking: A Survey", IEEE Trans. on Aero. and Electro. Sys., Vol. 34, No. 1, Jan 1998.
[12] Y. Bar-Shalom, K. C. Chang and H. A. P. Blom, "Tracking a Maneuvering Target Using Input Estimation Versus the Interacting Multiple Model Algorithm", IEEE Trans. on Aero. and Electro. Sys., Vol. AES-25, No. 2, Mar 1989.
[13] H. A. P. Blom and Y. Bar-Shalom, "The Interacting Multiple Model Algorithm for Systems with Markovian Switching Coefficients", IEEE Trans. on Automatic Control, Vol. 33, No. 8, Aug 1988.
[14] A. Munir and D. P. Atherton, " Maneuvering Target Tracking Using Different Turn Rate Models in the Interacting Multiple Model Algorithm", Proc. of the 34th Conf. on Decision and Control, Dec 1995.
[15] Kevin J. Bradshaw, Ian D. Reid and David W. Murry, " The Active Recovery of 3D Motion Trajectories and Their Use in Prediction", IEEE Trans. on Pattern Analysis and Mach. Intell., Vol. 19, No. 3, Mar 1997.
[16] Song De Ma, "A Self-Calibration Technique for Active Vision Systems", IEEE Trans. on Robotics and Automation, Vol. 12, No. 1, Feb 1996.
[17] Guo-Qing Wei, klaus Arbter and Gerd Hirzinger, "Active Self-Calibration of Robotic Eyes and Hand-Eye Relationships with Model Identification", IEEE Trans. on Robotics and Automation, Vol. 14, No. 1, Feb 1998.
[18] Chih-Yu Chen, ?D Target Trajectory Detection via Intelligent Monocular Visual Tracking System in Real-Time with Air-Target Orientation Recognition", Department of Electrical Engineering of National Taiwan University, Director: Professor Li-Chen Fu, 1998.