離散餘旋轉換之移動物體萃取
暨人種膚色分類

影像分割技術（Image segmentation）的研究，已經行之多年，是在許多影像處理的工作上扮演著非常基礎且關鍵的角色，例如物件的識別與辨識（Object recognition）、影像內容搜尋（Content-based image retrieval）以及視訊物件追蹤(Video Object Tracking)等等之許多應用，都必須將影像分割為具有某種意義的基本單位，例如前景后景的區分，移動中的物體與固定背景的分離，然後再將這些基本單位做更進一步有意義的處理，而image segmentation的基本單位可能是一個個的pixel或是block。在Part A中，Segmentation in video sequence based on DCT domain的工作，是將video sequence取出一張張的frame，將每張frame分割成8X8 pixels的block，然後經過一連的串運算來確定這個block是屬於前景或后景。主要的架構是將每個8X8的block轉成2 dimension-DCT domain(Discrete Cosine Transform)，獲得頻譜(frequency domain)上的訊息，再利用前后frame在同一個位置的block的互相關係，找出判別前后景的臨界值(threshold)，然後對此block進行偵測。此方法對於影片中移動物體的擷取有相當良好的表現，並且對于影片中的亮度變化(illumination variation)、背景移動物體(background moving object)有相當程度的低敏感性。 -隨著人機交互技術日益成長，人臉偵測問題越來越受到重視，成為當前研究的熱門領域，以及模式識別與電腦視覺領域研究的一個重點。而人臉膚色一直是人臉偵測的一個重要機制，皮膚顏色為人臉檢測提供了重要依據，膚色是人臉的重要訊息，具有相對的穩定性並且和大多數背景物體的顏色相區別。膚色不依賴面部細節特徵也不會隨著臉部表情或旋轉等變化而有所改變，因此針對彩色圖片利用膚色訊息進行快速檢測是人臉研究中的一項重要內容。在Part B中，提供了一種快速依據人類膚色辨別種族(race)的演算法，基本架構是將人種大約粗略分成白黃黑三種人種，再將大量的training image利用GMM(Gaussian Mixture Model)來尋找每一個人種的特徵值(feature parameter)，有平均值、標準差與權重(weighting)等，然後利用Bayesain Decision Rule來判定test image 所屬的人種。

關鍵字

移動物體偵測；膚色偵測

並列摘要

Research of image segmentation has been studied for many years. Image segmentation techniques are important but difficult in many image processing topics, such as object recognition and content-based image retrieval. In order to solve those problems, a successful image segmentation method is essential for splitting an image into meaningful regions, such as the discrimination between foreground and background of the segmentation of moving object and constant background. Then, make again these fundamental units (such as pixels or blocks) into further significant processing. In Part A, the task of Segmentation in video sequence based on DCT domain is dividing the video sequence into frames, and dividing 8X8 pixels element block each frame. Then passes through a continually string to calculate after determining this block is belongs to the foreground or the background. The main framework is to transfer each 8X8 block into 2 dimension-DCT domain in order to get the information of frequency domain, and then utilize the relation in the identical position block to calculate the threshold of background and foreground. This method has the quite good performance of catching moving object and also show to be low sensitive to illumination change and to noise. --------- With the growing technique of communication between human and robot, the problem of human face recognition has attached more importance and become the current research in the popular domain of computer vision and recognition model. Thus, human’s skin color is always an important mechanism and principle basis of human face detection. Human’s skin color has the relative stability with the difference of the majority background object appearance. The skin color does not rely on the face detail characteristic and do not change with the face expression and rotation. Therefore, utilizing skin color to examine human face in color image is an important context of human face recognition. In Part B, we provide a fast algorithm to identify human race with face skin color. The basic construction is roughly dividing human race into three parts: white, yellow and black race, then using Gaussian Mixture Model to train the feature parameter of each human race with large number of training images. Afterward, utilize Bayesian Decision Rule to determine the human race of test images.

並列關鍵字

moving object extraction ； skin color classification

參考文獻

[1]Jiashu Zhang, Like Zhang and Heng Ming Tai,”Efficient Video Object Segmentation using Adaptive Background Registration and Edge Based Change Detection Techniques” Volume 2, 27-30 June 2004 Page(s):1467 - 1470 Vol.2 IEEE Trans

[3]C Ridder, O Munkelt et al, “Adaptive Background estimation and foreground detection using kalman filtering”, Pro of Intl Conf On Recent Adbances in Mechatroincs(ICRAM),pp193-199,1995.

[4]Jesus Bescos, Jose M Menendez and Narciso Garcia”DCT based segmentation applied to a scalble zenithal people counter” IEEE 2003

[5]Mathieu Lamarre, James J Clark, ”Background subtraction using competing models in the block DCT domain” Volume 1, 11-15 Aug. 2002 Page(s):299 - 302 vol.1 Digital Object Identifier 10.1109/ICPR.2002.1044695 IEEE CNF

[6]L,Li and M Lenug, ”Integrating intensity and texture differences for robust change detection”IEEE Trans. On image processing, Vol 11,No2, pp105-112 2002

國際替代計量

離散餘旋轉換之移動物體萃取暨人種膚色分類

全文下載

主題瀏覽