Chin-Ya Huang

Mon-Ju Wu

ECE 533: Project Proposal

Image Segmentation in Sport

Purposed Topic

Image segmentation is an important technology for image processing.There are many applications whether on synthesis of the objects or computer graphic images requiring precise segmentation.With the consideration of the characteristics of eachobject composing images in MPEG4, object-based segmentation cannot be ignored.Nowadays, sports programs are among the most popular programs,and there is no doubt that viewers’ interest is concentrated onthe athletes. People most likely want to see more detail or clarityofthe athletes’ movements in images of a football game, theposture of an athlete who shoots the three point basket in an NBA tournament or the movement of a baseballin a major league game. Therefore, demand for image segmentation ofsport scenes is very high in terms of both visual compressionand image handling using extracted athletes.There are many algorithms used for image segmentation, and there doesn’t seemto be an optimal one proposed. According to class notes, we propose to segment images by using boundary extraction to get the shapes of athletes, and then compare the results by using the method in [3].The comparison is based on the execution time for a series frames, and the segmentation error probability. Since they may segment images automatically, they are useful for broadcasting during the games or establishing the video database on objects composing the image sequences.

Methods

To compare the difference of image segmentation by applying different methods, our approachesare as follows:

(1)Segment images by applying the concept of boundary extraction with the formula. Afterwards, we try to find a way to improve the precision of image segmentation.

(2)Segment images by applying the concept of [3]. We first calculate the color histogram to figure out the suitable HSI for image segmentation. Afterwards, we perform course image segmentation with the information of color histogram, and then apply marker watershed transform to get the fine image segmentation.

(3)Compare the results of image segmentation of applying different methods and compare the advantages and disadvantages of both methods.

References

[1] Class notes

[2] Text book.

[3] M. Naemura, A. Fukuda, Y. Mizutani, Y. Izumi, Y. Tanaka, and K. Enami,

“Morphological Segmentation of Sport Scenes using

Color Information, ” IEEE Transactions on broadcasting, vol. 46, no. 3, Sep. 2000.

[4] M. Tabb and N. Ahuja, “Multiscale Image Segmentation by

Integrated Edge and Region Detection, ” IEEE Transactions on image processing,

vol. 6, no. 5, May. 1997.

[5] F. Meyer, “Color image segmentation,” in Proc. Int. Conf. Image Processing,

Maastricht, The Netherlands, 1992.