以粒子群演算法實現
以粒子群演算法實現快速搜尋機器人移動位置之研究

近十年來，隨著視訊產品的普及化和電腦視覺技術快速發展，使得動態影像技術與追蹤方法被應用在各種領域，例如監視系統、停車場管理系統等。移動偵測與追蹤一直是電腦視覺技術相關研究的重要項目之一，傳統的機器人視覺系統一般都是採用影像處理的顏色和物件辨識，計算出受偵測物件的移動位置、速度和行動方向，在提供資訊給系統做決策，通常需耗費很多時間和空間複雜度運算。本論文使用影像相減和粒子群(Particle Swarm Optimization, PSO)演算法來建立基於電腦視覺之移動偵測及自動追蹤系統，找出辨識物件的最佳可能位置，以有效降低物件辨識時間、空間複雜度，使系統能更迅速做出適當戰略並正確控制機器人。此外，本研究使用一個適應性動態預估法，預估所要搜尋的範圍，去除不必要的運算量，實驗的結果顯示出，所提出的方法改進了多移動目標偵測的功能，且在明確目標的辨識效率上是相對的提高了。

關鍵字

足球機器人；多目標追蹤；粒子群演算法

並列摘要

In the last decade, with the popularization of video products and the rapid development of computer vision techniques, the detection and tracking methods for dynamic images have been widely applied in many fields, such as video surveillance, and parking area management systems. The motion detection and motion tracking has been one of the important computer vision technology related research projects. The traditional robot vision systems usually adopts color and object identification on image processing to determine the speed and action direction of the target, and then offer information for decision-making of the system However, it requires the highest computational load. In this research, the background subtraction and particle swarm optimization algorithms are used to build motion detection and motion automatic tracking system base on the computer vision, and find out the best possible position for identifying objects. The proposed approach can reduce the search space and computational complexity of the object identification, enable the system to make proper decision of strategy more rapidly, and control the soccer robot correctly. Furthermore, an Adaptive Search Range Function into search range, which reduces the unnecessary operation is proposed. In summary, the experimental results demonstrate that the proposed method improves the efficiency of multiple moving targets detection and recognizes target contour clearly.

並列關鍵字

robot ； multiple moving targets tracking ； particle swarm optimization

參考文獻

[7] 王榮彬(民95)，高效能即時多物件移動偵測追蹤系統，長榮大學經營管理研究所碩士論文。

[3] 陳寬益(民91)，即時視覺伺服器追蹤設計系統之設計與實現，成功大學機械工程學系碩士論文。

[9] A.C.K.Ng, and B.Zeng, “A new fast motion estimation algorithm based on search window sub-sampling and object boundary pixel block matchinh,”Proc. ICIP 98, Oct.1998, pp 605-608.

[10] B. Liu, and J. Zhang, "Adaptive Tracing System with a Spatial Light Modulator," Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on, vol. 1, June 2006, pp 5111 - 5114.

[11] C.H. Kung, C.M. Kung and J.P. Wang, “High Performance Real-time Object Detecting and Tracking System for Multiple Moving Targets” IEEE Instrumentation and Measurement Technology Conference, May 2007, pp 1-4.

國際替代計量

以粒子群演算法實現以粒子群演算法實現快速搜尋機器人移動位置之研究

全文下載

主題瀏覽