Codebook影像技術應用於即時背景影像重建及移動物件分離之研究

本論文研究目的 (一)為背景重建，在有移動物件干擾及動態場景的干擾下，重建出完整的背景；(二)為移動物件分離，在有光影變化、樹葉晃動及看板燈光的干擾下，能夠較完整的擷取出移動物件。這兩種影像技術應用在數位影像處理中，能夠應用的範在圍很多，可以作為判斷事件的良好依據，例如：風景區中的人群和街景區的行人等移動物體追蹤、監控系統偵測、物件辨識…等。本研究以移動物影像擷取及背景重建為技術主旨，移動物件偵測的部份採取codebook 背景模型，利用影像序列中的顏色和亮度失真，做為每一個像素點來形成一個codebook，也就是前景部份，再來分為兩部份，一部份是針對移動物件的濾除，對每張影像且每一點RGB像素值做比對，統計出該場合下屬於背景的RGB像素值，再利用與原始影像做平均，設定權重值取得平均影像，統計出來的背景與平均影像結合，建構出乾淨的背景影像。另一部份是保留移動物件，對另一張背景影像進行像素點掃描，而進行更換不同背景的場景影像。

關鍵字

codebook背景模型；背景重建；更換場景

並列摘要

The aim of this study to investigate two digital image processing techniques：(a)background reconstruction：in the interference of a motion object interference and dynamic scenes, the reconstruction of a complete background.(b)separation of motion object：in the interference of light and shadow changes, leaves swaying and billboard lighting for a more complete capture of motion objects. The application of these two purposes can be used in digital image processing, to a broader scope and can be used as a on basis to determine the events, such as, people in scenic area and street scenes, mobile object tracking, monitoring system detection, object recognition, etc. This study focus on the image capture of a motion object and background reconstruction technology. Motion object detection uses part of the codebook background model： color and brightness distortion in the image sequence, each pixel forms a codebook, which is part of the foreground. Then, dividing into two parts. One part is filtering mobile objects. Every point of the RGB pixel values for each images is analyzed and compared for the RGB pixel values of the background. Then, averaging the original image, set the weight values to obtain the average image and calculate the background combined with the average image to construct a clean background image. The other part is to keep motion objects on another background image pixel scan for switching to a different background scene images.

並列關鍵字

codebook background model ； background reconstruction ； replacement of the scene

參考文獻

[2] Nicholas A. Mandellos, Iphigenia Keramitsoglou and Chris T. Kiranoudis, “A Background Subtraction Algorithm for Detecting and Tracking Vehicles,” E-Product E-Service and E-Entertainment (ICEEE), 2010.

[4] A. J. Lipton, H. Fujiyoshi, and R. S. Patil, “Moving target classification and tracking from real-time video,” IEEE Workshop Applications of Computer Vision, 1998, pp. 8-14.

[5] R. Cutler and L. S. Davis, “Robust real-time periodic motion detection, analysis, and applications,” IEEE Trans. Pattern Anal. Machine Intell., vol. 22, Aug. 2000, pp. 781-796.

[6] S. C. Cheung and C. Kamath, “Robust techniques for back ground subtraction in Urban Traffic Video,” Proceedings of SPIE, 2004.

[7] S. Huwer and H. Niemann, “Adaptive Change Detection for Real-time Surveillance Applications,” Proceedings of IEEE third International Workshop on Visual Surveillance, July 2000, pp. 37-46.

國際替代計量

Codebook影像技術應用於即時背景影像重建及移動物件分離之研究

未授權

主題瀏覽